[go: up one dir, main page]
More Web Proxy on the site http://driver.im/

CN1823298A - Object decision device and imaging device - Google Patents

Object decision device and imaging device Download PDF

Info

Publication number
CN1823298A
CN1823298A CN200480020302.XA CN200480020302A CN1823298A CN 1823298 A CN1823298 A CN 1823298A CN 200480020302 A CN200480020302 A CN 200480020302A CN 1823298 A CN1823298 A CN 1823298A
Authority
CN
China
Prior art keywords
face
people
decision
unit
image
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN200480020302.XA
Other languages
Chinese (zh)
Other versions
CN100538498C (en
Inventor
井尻善久
露口卓也
今江文一
山元昌史
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Omron Corp
Original Assignee
Omron Corp
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Omron Corp filed Critical Omron Corp
Publication of CN1823298A publication Critical patent/CN1823298A/en
Application granted granted Critical
Publication of CN100538498C publication Critical patent/CN100538498C/en
Anticipated expiration legal-status Critical
Expired - Fee Related legal-status Critical Current

Links

Images

Landscapes

  • Studio Devices (AREA)

Abstract

Face detection is executed only on an image imaged based on an arbitrary (predetermined) focal point information defined in advance. In face detection, the face of a person is detected based on a relative value of statistics in a plurality of characteristic regions produced by contour or parts of the face of the person. Thus, even if the face in the image used for face detection is blurred, the statistics in a certain region can be acquired, and thus such face is detectable. Therefore, the primary subject to be focused can be rapidly detected without performing focal point control at a stage of preliminary imaging in the imaging apparatus.

Description

Object determination device and camera head
Technical field
The technology of device (digital camera (digital still camera) and digital camera (digital video camera) etc.) of still image and dynamic image etc. the present invention relates to effectively to be applied to make a video recording.
Background technology
Automatic focus that specific subject is focused and the technology such as automatic exposure control of carrying out exposure control according to specific subject in camera head, be equipped with.In the past, the subject that the ad-hoc location of predesignating in frame (for example central point in the frame) is existed was implemented automatic focus or automatic exposure control.Yet according to user's framing (framing) hobby, the user wants the subject paid close attention to, wants that promptly the subject as the object of automatic focus or automatic exposure control is not limited to be positioned at the central authorities of frame.In this case, the user operates camera head so that the subject of wanting to pay close attention to moves to the ad-hoc location in the frame, after having carried out automatic focus or automatic exposure control under this condition, must implement the framing corresponding with hobby, carrying out this processing is the operation that extremely bothers.
As the technology that addresses this problem, the technology of implementing shooting (following this shooting is called " preparation shooting ") for the object determining to focus or exposure controlling object is in advance arranged.In preparing the technology of shooting, use the image of in the preparation shooting, being made a video recording to determine the object of focusing or exposure control.
As the concrete example of the technology of preparing shooting, the technology (with reference to patent documentation 1) of irrespectively face of the personage in the image being focused with position/scope of the personage of visual field (image in) is arranged.Equally, have with image in position/scope of personage irrespectively the personage in the image is carried out the technology (with reference to patent documentation 2) of proper exposure.
In above-mentioned prior art, when in picture, detecting a plurality of face, apart from least a portion of the near face of central portion as the ranging region control of focusing automatically.And the technology as under the situation that detects a plurality of faces equally in picture also has the technology (with reference to patent documentation 3) that the personnel that automatically make in the picture enter the depth of field.
Patent documentation 1: the spy opens the 2003-107335 communique
Patent documentation 2: the spy opens the 2003-107555 communique
Patent documentation 3: the spy opens the 2003-092699 communique
Summary of the invention
Yet, in prior art (for example patent documentation 1~3 described technology), have following problem.That is, the problem of existence is, if initial focus significantly departs from respect to the main objects body in the image that obtains by the preparation shooting, then exists and can not detect the possibility that the main objects body is a face.
As the method that addresses this problem, the method by the image that carries out focal point control in the preparation shooting stage, make a video recording detecting the main objects body is arranged.Specifically, the method that detects the main objects body from a plurality of images that obtain by the preparation shooting is for example arranged.In the method, a plurality of images that carried out focusing of shooting when the preparation shooting.Then, implement the face detection of main objects body according to these a plurality of images.Therefore, even in certain image because under the fuzzy and non-detectable situation of face of main objects body, also can in other image, detect the face of main objects body.
Yet,, must carry out face to a plurality of images and detect in this method.And,, be necessary to carry out focal adjustments for a plurality of images of having focused of shooting in camera head.Therefore, compare, exist focusing process to need the problem of time with existing method.
In the present invention, first purpose is to address this problem, and does not carry out focal point control in the preparation shooting stage in camera head, and fast detecting becomes the main objects body of the object of focusing or exposure control.
In addition, deciding under the situation of object of focusing or exposure control, there is following problem by carrying out above-mentioned preparation shooting.Not necessarily one of the number of detected people's face in the image of preparation shooting.Be under one the situation at the number of detected people's face, as long as unconditionally this face is focused or expose control.Yet, in the image of preparation shooting, detect under a plurality of personages' the situation of face, tackle which face is focused or the control that exposes, took into full account as yet up to now.For example, have according to being the policy of user expectation, the nearest people's face of distance frame central authorities is focused or the technology of the control that exposes as the face of subject at the captured people's face of frame central authorities.Yet this technology is to expect that based on the user face as subject is such prerequisite, if there be the face of the equal expectation of a plurality of quilts as subject in frame, can not tackle reliably most in frame.
In the present invention, second purpose is to address this problem, and is provided under the situation of the face that detects a plurality of personages in the preparation shooting, judges which face of reply is focused or the device of the control that exposes.
(first mode)
In the following description, the colour of skin means the proprietary colour of skin, is not limited to the colour of skin of specific ethnic group.
In order to reach above-mentioned first purpose, the present invention adopts following structure.First mode of the present invention is a kind of object determination device, comprises face detecting unit and decision unit.
The face detecting unit detects people's face to the image of the image of making a video recording according to focus information arbitrarily as process object from the image of aforementioned process object.Focus information can be the focus information of regulation arbitrarily.Here, the focus information of regulation is to be decided to be one focus information in advance, rather than the focus information of dynamic change.Therefore, come the device of photographed images at focus information according to the rules, promptly the shooting be imported in the camera head of the image in the face detecting unit, expectation be the focal point control of implementing in advance based on the focus information of this regulation.For example, expectation is to implement the first-class control of focus lamp in advance according to this focus information.
By controlling camera head so in advance, just there is no need when shooting is imported into image in the face detecting unit, to implement again focal point control.Therefore, shooting is imported into the needed time shortening of image in the face detecting unit.
Yet, in the image that focus information is so according to the rules made a video recording, have the possibility of the subject in the image out of focus.Therefore, in the image in being imported into the face detecting unit, exist and answer the image blurring of detected face, the possibility of being made a video recording originally unintelligiblely.
Therefore, the face detecting unit that comprises of first mode of the present invention constitutes the image that also detects the face of being made a video recording unintelligiblely.Specifically, the face detecting unit detects people's face according to the relative value because of the statistic of the concavo-convex of people's face or a plurality of characteristic areas that organ produces.That is, the face detecting unit that first mode of the present invention comprises is not the shape that constitutes according to by the edge in the image etc., but carries out the detection of face according to the relative value of the statistic of characteristic area.The statistic of characteristic area is the value of the total of the pixel value in the characteristic area, average, distribution etc.Relative value is for example than the value of, difference, statistical distance, cross-entropy etc.And, also can carry out face and detect according to the relative value of the deep or light value of characteristic area.
In the image of being made a video recording, be difficult to discern the shape that constitutes by edge etc. unintelligiblely.Yet, about the statistic of certain characteristic area, even for example the image of being made a video recording also can be obtained unintelligiblely.Therefore, the face detecting unit that comprises of first mode of the present invention can detect people's face from the image (image of process object) that focus information is according to the rules made a video recording.
The decision unit determines subject of focusing and/or the subject that becomes the object of the control that exposes according to by the detected face of face detecting unit when the enforcement shooting.The subject of focusing when implementing shooting can be by the detected face of face detecting unit itself with the subject of the object that becomes the control that exposes, and also can be each position (eye, nose, mouth, ear, volume etc.) that constitutes detected face.
In the object determination device in first mode of the present invention that constitutes like this, from the image of being made a video recording, detect people's face, according to the subject of detected face decision becoming focusing object or become the subject of the object of the control that exposes according to focus information (focus information of regulation) arbitrarily.As mentioned above, camera head is carried out focal point control, the image that according to the rules the focus information of can making a video recording is apace made a video recording by according to the rules focus information in advance.And, from this image, detecting people's face, decision becomes the subject of focusing object or becomes the subject of the object of the control that exposes.
Therefore, according to the object determination device in first mode of the present invention, in camera head, the shooting stage do not carry out focal point control, can detect apace and determine to become and focus or the main objects body (position of face or face etc.) of the object of the control that exposes in preparation.
And, according to the object determination device in first mode of the present invention, even under the situation that constitutes the image of in camera head, obtaining many preparation shootings, also can from preparation photographed images still less, detect face.Therefore, can cut down preparation needed time of shooting and face and detect the needed time.
Face detecting unit in first mode of the present invention can constitute by the following stated and detect people's face.That is, at first in the image of being made a video recording, determine region-of-interest.Region-of-interest is and the identical zone or the littler zone of image size of being made a video recording.
Then, to obtain the characteristic area in this region-of-interest be the relative value of the statistic of first area and second area to the face detecting unit.First area and second area are respectively the zoness of different in the region-of-interest.
Then, the face detecting unit judges whether include people's face in region-of-interest according to the relative value of the statistic of being obtained.The face detecting unit can constitute by for example the relative value of certain threshold value and statistic being compared and carry out this judgement.And the face detecting unit also can constitute by with reference to the relative value of statistic be illustrated in the table that the value that whether includes people's face in the region-of-interest is mapped and stores under the situation of the relative value of calculating certain statistic, carries out this judgement.And the face detecting unit is judged according to this and is detected people's face.
Decision unit in first mode of the present invention also can constitute at the face detecting unit and detect under the situation of a plurality of faces, decides the subject of focusing or becomes the subject of object of the control that exposes by the following stated.That is, the decision unit is at first according to determining a face by the detected a plurality of faces image information separately of face detecting unit.The image information of face is represented the information that obtained from the image of face, for example be face towards the size of the image of, face, the position, the density etc. of face on every side of face in the image.
The density of face is on every side represented by the number of the face that for example (certain peripheral region) exists around detected face certain.If should be worth greatly, then there are many faces around, for example represent that the people of detected face is comprised in the collective.On the other hand, if should value little, then the face that exists around be few, for example is illustrated in the place that the people of detected face separates with collective, exists independent or the several people.
And the decision unit is according to the face that is determined, the decision subject of focusing and/or become the subject of the object of the control that exposes.
In first mode of the present invention that constitutes like this, according to the image information of a plurality of faces, judgement decides the subject of focusing according to which face and/or becomes the subject of the object of the control that exposes.Therefore, even detected by the face detecting unit under the situation of a plurality of faces, the user also need not implement to select, and selects a face by the decision unit from a plurality of faces.
Decision unit in first mode of the present invention can constitute a plurality of faces image size is separately compared, and decides the subject of focusing for maximum face and/or becomes the subject of the object of the control that exposes according to its size.
The face of the maximum image in a plurality of faces is generally in the nearest position of distance camera head.Therefore, first mode of the present invention according to such formation, the user need not implement the selection of face, is determined but be positioned at apart from the face of the nearest position of camera head, decides the subject of focusing or becomes the subject of object of the control that exposes according to this face.
Owing to the object in front of camera head is positioned at, image is just fuzzy more, thereby in general must focus at the object in front of camera head is positioned at.And,, comprise the user's that makes a video recording main objects body or be exactly possibility (below show as " importance of the face ") height of main objects body itself face and the appearance ratio that is positioned at the back of camera head in front of being positioned at.In this, be effective according to being positioned at the subject of object that decides the subject of focusing and/or become the control that exposes apart from the face of the nearest position of camera head.
Decision unit in first mode of the present invention also can constitute at each face, according to any or the multinomial score of calculating in the density of the size of, face, position in image or face on every side of face.At this moment, the decision unit is used as image information to this score of calculating.
In the object determination device in first mode of the present invention that constitutes like this, wait according to the density towards the size of, face, position in image or face on every side of face and to estimate each face.For example, face towards being the sideways face height of importance specific surface that is be evaluated as its face towards the face in front.And for example, the face that the size of face is bigger is be evaluated as the importance of face than little face height.And for example, the position in image is to be evaluated as the importance of face than the face height that is positioned at away from the zone of central authorities near central face.And the face that the density of face on every side is high is be evaluated as the low density face height of the importance of face than face on every side.Yet the height of this evaluation can be set arbitrarily by deviser or user.
According to first mode of the present invention of such formation, set according to the height of this evaluation, selecting the score height is the high face of importance.
Decision unit in first mode of the present invention also can constitute at the face detecting unit and detect under the situation of a plurality of people's faces, decides the subject of focusing and/or becomes the subject of the object of exposure control by the following stated.That is, the decision unit is according to the result to the gained of being found range respectively by the material object of the detected a plurality of faces of face detecting unit, the decision subject of focusing and/or become the subject of the object of the control that exposes.This range finding can be implemented by the range cells or the distance measuring equipment that are possessed in the camera head for example, also can be implemented by the range cells that is possessed in the object determination device.Yet this range finding be owing to must measure the distance of the material object of camera head and face, if thereby camera head and object determination device be separated, can not carry out the range finding of implementing by the range cells that is possessed in the object determination device.
Range finding realizes (active mode) by for example material object ejaculation infrared ray to face.And range finding also can be used the mode (for example passive mode) beyond this active mode.And the decision unit decides the subject of focusing according to its distance for minimum face and/or becomes the subject of the object of the control that exposes.Like this, decide the subject of focusing according to the distance of the material object of camera head and face for minimum face or become the control that exposes object subject effect as mentioned above.
(second mode)
Second mode of the present invention is a kind of camera head, comprises that image unit, face detecting unit, decision unit, information obtain unit and image storage unit.
The image unit shooting obtains the obtained focus information in unit and/or the image of the control information that exposes based on the focus information of regulation or by information.The focus information of regulation is that to be predetermined be one focus information, rather than the focus information of dynamic change.On the other hand, obtaining the obtained focus information in unit by information is not the focus information that is predetermined, but the focus information of dynamic change.More particularly, be to obtain the focus information that the unit determines by face detecting unit, decision unit and information.That is, be to be used for deciding the subject of focusing according to detected face, to the focus information of the subject focusing that this determined,
Under the situation of the focus information of not obtaining unit output by information, image unit is controlled so as to the state that can obtain based on the image of the focus information of stipulating.And, when the shooting indication, obtain image based on the focus information of regulation.On the other hand, have under the situation of focus information or exposure control information obtaining unit output by information, image unit is controlled so as to the state that can obtain based on the image of these information.And when the shooting indication, image unit is obtained the image based on this focus information or exposure control information.
The face detecting unit detects people's face according to the relative value because of the statistic of the concavo-convex of people's face or a plurality of characteristic areas that organ produces from the image of process object based on the focus information of regulation.The decision unit determines subject of focusing and/or the subject that becomes the object of the control that exposes according to by the detected face of face detecting unit when the enforcement shooting.The face detecting unit of second mode of the present invention with the decision unit have basically with first mode of the present invention in the face detecting unit and the decision identical structure and the effect in unit.
Information obtain the unit obtain be used for to by the focus information of decision subject that the unit determined focusing and/or be used for by the decision subject that the unit determined as the expose exposure control information of control of object.
Image storage unit stores by image unit according to obtain obtained focus information in unit and/or the exposure control information image of making a video recording by information.That is, obtained focus information or exposure during control information when obtained the unit by information, image unit is implemented focal point control or exposure control according to this focus information or exposure control information.By this focal point control, to focusing by the decision subject that the unit determined.And,, implement and control by the corresponding exposure of the decision subject that the unit determined by this exposure control.Then, image unit is carried out shooting, the data of the image that image storage unit stores is made a video recording.
And second mode of the present invention can constitute also has range cells, and it is found range to the distance from camera head to any subject.Yet, obtain the unit in information and constitute by the distance of the material object of subject and camera head is found range and obtain under the situation of focus information, can possess information and obtain the substituting unit of unit as range cells.That is, in this case, information obtains the unit and moves as range cells.
Constituting like this under the situation of second mode of the present invention, the decision unit constitutes at the face detecting unit and detects under the situation of a plurality of people's faces,, decide the subject of focusing for minimum face and/or become the subject of the object of the control that exposes find range the respectively result of gained of the material object of a plurality of faces according to range cells according to its distance.
In second mode of the present invention, in camera head, be equipped with object determination device as first mode of the present invention.Therefore, as camera head, can obtain and the identical effect of first mode of the present invention.
(Third Way)
Third Way of the present invention is a kind of object determining method, is the method that comprises following each step.At first, signal conditioning package is from according to the image that arbitrarily focus information is made a video recording, and detects people's face according to the relative value because of the statistic of the concavo-convex of people's face or a plurality of characteristic areas that organ produces.Then, signal conditioning package determines subject of focusing and/or the subject that becomes the object of the control that exposes according to detected face when the enforcement shooting.The Third Way of the present invention that constitutes like this obtains and the identical effect of first mode of the present invention.
(cubic formula)
Cubic formula of the present invention is a kind of image capture method, is the method that comprises following each step.Below, enumerate this each step.Signal conditioning package indication camera head is to obtaining the step that the image of the obtained focus information in unit is made a video recording based on the focus information of regulation or by information.Signal conditioning package detects the detection step of people's face according to the relative value because of the statistic of the concavo-convex of people's face or a plurality of characteristic areas that organ produces from the image of being made a video recording by camera head focus information according to the rules.Signal conditioning package determines the deciding step of the subject of subject of focusing and/or the object that becomes the control that exposes according to detected face when the enforcement shooting.Signal conditioning package is obtained and is used for the focus information that the subject that is determined is focused and/or is used for the subject that is determined as the expose step of exposure control information of control of object.The step of the image that signal conditioning package indication memory device stores is made a video recording according to obtained focus information and/or exposure control information by camera head.
And cubic formula of the present invention can constitute and also comprise the step that signal conditioning package indication distance measuring equipment is found range to the distance from camera head to any subject.In this case, if in detecting step, detect a plurality of people's faces, then signal conditioning package is in deciding step,, decide the subject that carries out aforementioned focusing for minimum face and/or become the subject of the object of the control that exposes find range the respectively result of gained of the material object of a plurality of faces according to distance measuring equipment according to its distance.
In the cubic formula of the present invention that constitutes like this, the object determining method in the Third Way of the present invention is carried out as the step that is included in the image capture method.Therefore, in cubic formula of the present invention, can obtain the effect identical with Third Way of the present invention.
(the 5th mode)
The 5th mode of the present invention is a kind of object determination device, comprises detecting unit and decision unit.Detecting unit detects the subject of the colour of skin with people from the basis image that focus information is made a video recording arbitrarily.As the example of the subject of the colour of skin, face, hand, health of people etc. are arranged with people.By such formation detecting unit,, handle the needed time yet can cut down to detect although can not detect personage's face exactly.And, also can cut down the needed time of installation and the cost of detecting unit.
The decision unit determines subject of focusing and/or the subject that becomes the object of the control that exposes according to by the detected subject of detecting unit when the enforcement shooting.
In the 5th mode, detecting unit detects the subject of the colour of skin with people.That is, the included detecting unit of the 5th mode of the present invention is the same with first mode of the present invention, does not carry out the detection based on the shape that is made of edge in the image etc.
In the image of being made a video recording, be difficult to discern the shape that constitutes by edge etc. unintelligiblely.Yet, even from the image of for example being made a video recording, also can detect subject unintelligiblely with certain particular color.Therefore, the included detecting unit of the 5th mode of the present invention can detect the subject that has the possibility that is people's face, the subject that promptly has people's the colour of skin from the image that focus information is according to the rules made a video recording.
(distortion of first mode)
First mode of the present invention can constitute and comprise that also frame obtains the unit, and this frame is obtained the unit from according to the image of the frame of obtaining regulation the sequential chart picture that comprises a plurality of frames that focus information is made a video recording the arbitrarily image as process object.Under situation about constituting like this, the face detecting unit in first mode the image of the frame of obtaining the obtained regulation in unit by frame as process object.
In first mode of the present invention that constitutes like this,, can obtain and the identical effect/effect of first mode of the present invention for the sequential chart picture that comprises a plurality of frames.
And, decision unit in first mode of the present invention can constitute when being detected a plurality of people's face by the face detecting unit, according to the position of these a plurality of people's faces, from these a plurality of people's faces, determine the face that when implementing shooting, becomes the face of the object of focusing and/or become the object of the control that exposes.By such formation, can obtain and the identical effect/effect of the 6th mode of the present invention described later.
And, decision unit in first mode of the present invention can constitute when being detected a plurality of people's face by the face detecting unit, is benchmark the people's face decision that is positioned at its center for becoming the face of the object of focusing and/or become the face of the object of the control that exposes when implementing shooting with the number of detected face.By such formation, can obtain and the identical effect/effect of the 7th mode of the present invention described later.
And the decision unit in first mode of the present invention can constitute and comprise: taxon, and when being detected a plurality of people's face by the face detecting unit, its detected a plurality of people's faces are categorized as a plurality of set; Tentative decision unit, it determines the face that becomes the face of the object of focusing and/or become the object of the control that exposes when implementing shooting at each set in a plurality of set people's face that is comprised in this set; And the final decision unit, it is from being become the face of object by final decision the tentative decision face that the unit determined.By such formation, can obtain and the identical effect/effect of all directions formula of the present invention described later.
And, first mode of the present invention can constitute and also comprise decision unit, district, its testing result according to the face detecting unit decides the face of selecting district, this selections district to be used to determine to become the face of the object of focusing when implementing to make a video recording and/or becoming the object controlled of exposing.Under such situation about constituting, the decision unit constitutes the face that determines to become object in selecting the district the face that is comprised.By such formation, can obtain and the identical effect/effect of the 9th mode of the present invention described later.
And first mode of the present invention can constitute also and comprise: judging unit, and when being detected a plurality of people's face by the face detecting unit, it judges the face of the maximum in detected people's face; And selected cell, it is selected maximum face and has the face of the size in the specialized range that size with this maximum face is a benchmark from detected face.Under situation about constituting like this, the decision unit constitutes from selected face decision becomes the face of the object of focusing and/or becomes the object of the control that exposes when implementing shooting face.By such formation, can obtain and the identical effect/effect of the tenth mode of the present invention described later.
And first mode of the present invention can constitute also and comprise: taxon when being detected a plurality of people's face by aforementioned face detecting unit, is categorized as a plurality of set with detected a plurality of people's faces according to separately position; And set decision unit, it determines to select set from a plurality of set, and this choice set share the face that becomes the face of the object of focusing and/or become the object of the control that exposes when implementing shooting in determining.Under situation about constituting like this, the decision unit constitutes from become the face of object by decision the face that is comprised in the set decision selection that the unit the determined set.
(distortion of second mode)
Second mode of the present invention can constitute the sequential chart picture that the image unit shooting comprises a plurality of frames.Second mode of the present invention constitutes and comprises that also frame obtains the unit under situation about constituting like this, and it obtains the image of the frame of regulation from the sequential chart picture that comprises a plurality of frames of being made a video recording by image unit.And the face detecting unit constitutes and detect people's face from is obtained the obtained image in unit by frame.And image storage unit constitutes the sequential chart picture that storage is made a video recording by image unit.
In second mode of the present invention that constitutes like this,, can obtain and the identical effect/effect of second mode of the present invention for the sequential chart picture that comprises a plurality of frames.
And second mode of the present invention can constitute and also comprise display unit, and it distinguishes the demonstration of coming by decision people's face that the unit determined and other face.Display unit for example frame be presented at people's face of being determined around, thereby with other face difference demonstration of coming.In addition, display unit is for example showing around the people's face that is determined that the frame with shown around other face has the frame of different colors, thickness or shape, thereby distinguishes the demonstration of coming with other face.In addition, display unit is for example implemented the Flame Image Process different with other face to the people's face that is determined and is shown, thereby distinguishes the demonstration of coming with other face.Display unit is by carrying out this demonstration, which face the user can easily be judged according to decide the subject that becomes the focusing object etc.
(the 6th mode)
In order to reach above-mentioned second purpose, the present invention adopts following structure.The 6th mode of the present invention is a kind of object determination device, comprises detecting unit and decision unit.The 6th mode of the present invention is according to the position of the face that is detected, and decision becomes the face of the object of the control of should focusing and expose.In addition, the face that is determined by the 6th mode of the present invention can be used as the object of any processing, and is not limited to focusing and exposure control.For example, can be used as the object of processing such as carrying out color correction (white balance correction), emphasizing contour.
Specifically, detecting unit detects people's face from the image of being imported.Detecting unit can be used existing any face detection tech so long as can detect the technology of a plurality of people's faces from image.For example, can detect have the colour of skin the zone as the face zone.As other example, the detection technique of template at the position of using face or face and the technology etc. that detects the face zone according to deep or light difference are arranged.
When detecting unit detected a plurality of people's face, the decision unit was according to the position of this a plurality of people's faces, and decision becomes the face of the object of focusing and/or becomes the face of the object controlled of exposing when implementing shooting from these a plurality of people's faces.The decision unit is not to consider that by independent the position of each face determines, but determines according to the position of a plurality of faces.That is, the decision unit determines according to the relative position relation of a plurality of faces at least.The decision unit can be regarded the position of face as the position of the point that face comprises, and also can see the position in the zone of wining honour for.
In addition, can constitute focusing by the part (for example eye, nose, volume, mouth, ear) of the decision face that the unit determined or exposing control.
According to the 6th mode of the present invention, according to the position of detected a plurality of faces, decision becomes the face of the object of focusing or exposure control.Therefore, even under the situation that from carried out image that the preparation shooting obtains, detects the face of Duoing, also can determine the face of defined amount than the number of the face of the object that should become focusing or exposure control.
And, not independently to consider at the position of each detected face, but according to the face of the object of the relative position of a plurality of faces decision becoming focusing or exposure control.Therefore, decide the face of the object that becomes focusing or exposure control according to the situation of detected a plurality of faces (for example dense degree of face, position that face is intensive).Therefore, can carry out focusing process corresponding or exposure control with the situation of detected a plurality of faces.
Below, the more specifically structure of the decision unit in the 6th mode of the present invention is described.Decision unit in the 6th mode of the present invention can constitute and comprise decision unit, center and face decision unit.
Decision unit, center determines the center of the position of these a plurality of people's faces according to the position of detected a plurality of people's faces in detecting unit.Here said central representation conceptual Center.For example, be called under the situation at center a plurality of conceptual Centers such as expression unfaithful intention, heart, orthocenter, center of gravity at certain 3.Like this, here, the center that said central representation is obtained according to certain notion.
Specifically, decision unit in center can constitute the polygonal center external with the position of a plurality of people's faces of obtaining.At this moment, polygon can be the polygon with summit of the number that is predetermined.In this case, the polygon that is generated there is no need necessarily to have the somebody of institute face as the summit.
For example, be under the leg-of-mutton situation at external polygon, in the circumcentre of a triangle obtained, heart, orthocenter, the center of gravity any one obtained as the center in decision unit in center.
And, for example, be under the tetragonal situation at external polygon, the tetragonal intersection of diagonal obtained is obtained as the center in decision unit in center.And, for example, be under the tetragonal situation at external polygon, decision unit, center can constitute this quadrilateral is divided into two triangles, obtain in unfaithful intention, heart, orthocenter, the center of gravity any one at each triangle respectively, obtain the center according to obtained 2.
And, for example, external polygonal number of vertex be more than or equal to 5 situation under, decision unit, center can constitute external polygon is divided into a plurality of triangles, obtain in unfaithful intention, heart, orthocenter, the center of gravity any one at each triangle, use obtained point to constitute new polygon, obtain the center by repeating above-mentioned processing at this polygon.
And decision unit, center can constitute the decision of the center of gravity of the position of a plurality of people's faces is the center.
Below, determine the structure of unit to describe to face.Face decision unit is according to the position at the center of being obtained by decision unit, center, and decision becomes the face of object.Face decision unit is according to the face that becomes the focusing object when implementing shooting and/or become the number of face of the object of the control that exposes, and decision becomes the face of object.
Specifically, face decision unit can constitute being positioned at the people face decision nearest apart from the center for becoming the face of object.
And face decision unit can constitute from being arranged in apart from the face of center gauge set a distance, the face of benchmark decision becoming object according to the rules.The benchmark of regulation for example can be based on the benchmark (benchmark of the face of for example decision maximum of the size of face, the benchmark of the face of size in the middle of the decision, the benchmark of the face that decision is minimum), can be based on the benchmark benchmark of the nearest face of image central authorities (for example apart from) of the position of the face in the image, can be based on face towards the benchmark benchmark of the face in front (for example towards), can be based on the benchmark (amount that for example determines expression to be similar to the degree of face is the benchmark of the face of maximum) that is similar to face, the benchmark that can be based on the sex of inferring according to face (for example determines to be speculated as the benchmark of the male sex's face, decision is speculated as the benchmark of women's face), (decision is speculated as the benchmark of face in minimus age can be based on the benchmark in the age of inferring according to face, decision is by the benchmark of the face in age in the middle of being speculated as), also can be other any benchmark.And the benchmark of regulation also can be that said reference is suitably carried out benchmark after a plurality of combinations.
In the 6th mode of the present invention that constitutes like this,,, decide the face of the object that becomes focusing or exposure control according to the center of the position of a plurality of faces as the benchmark of the relative position of a plurality of faces.For example, the face that is positioned at the approximate center of a plurality of faces is decided to be the face of the object that should become focusing or exposure control.Under the situation that collective is photographed (situation that a plurality of faces are intensive), the most cases user focuses or exposes control near the face (unconsciously) the central authorities of this collective.Therefore, by this user's of automatic realization manual operation, can save user's time.Therefore, in automatic focus or automatic exposure control, general expectation be that near the face the central authorities of a plurality of faces of being arranged in collective is focused or exposed control.And, according to the 6th mode of the present invention of above-mentioned formation, can carry out the control of above-mentioned expectation automatically, can save user's time.
And the decision unit in the 6th mode of the present invention can constitute the face decision that is positioned at below in a plurality of people's faces to becoming the face of aforementioned object.In addition, to there is no need be strictly to be positioned at the face of below to the face of below, can be the face that for example is positioned at from beginning the 2nd down, also can be the face that is positioned at from beginning the 3rd down.Gather under the situation of taking pictures under the state that people's face is vertically arranged, the face that generally is positioned at the below is apart from the near more face of the distance of camera head.For example, squat down front-seat people, back row people stands under the situation of taking pictures, and just becomes above-mentioned state.And, for example,, just become above-mentioned state using high platform to gather under the situation of taking pictures.Therefore, by the face of below is considered as focusing or the face of the object of the control that exposes being positioned at, the possibility of the nearest face of distance camera head being carried out focusing or exposure control becomes big.
The 6th mode of the present invention can constitute and also comprise taxon and set decision unit.Taxon is categorized as a plurality of set with detected a plurality of people's faces according to separately position when being detected a plurality of people's face by detecting unit.More particularly, taxon is categorized as a plurality of set to detected a plurality of faces, so that the close face in the position in the image is comprised in the identity set by using general sorting technique.
Set decision unit determine (selections) to be used to determine to become the set (that is, comprising the set of the face of the object that becomes focusing or exposure control: be called " selection is gathered ") of face of the object of focusing or exposure control from a plurality of set of being classified by taxon.That is, from become the face of the object of focusing or exposure control by final decision the face that is comprised in the set decision selection that the unit the determined set.
Set decision unit according to which kind of benchmark decides is selected set all can.For example, set can be selected according to the number decision of the face that is comprised in each set in set decision unit, set is selected in the center decision of the position of a plurality of faces that comprised in can gathering according to each, can be according to (for example based on the obtained characteristic quantity of the face that comprised in each set, based on the obtained amounts such as number towards, face of the size of face, face, the frequency in the frequency of the particular sex that expression is inferred according to face, generation year of inferring according to face or be similar to the amount of the degree of face) decision selects set.
Constitute under the situation that comprises above-mentioned taxon and set decision unit in the 6th mode of the present invention, the decision unit constitutes the people's face that is comprised in the set according to selecting, and decision becomes the face of object the people's face that is comprised in selecting set.
According to the 6th mode of the present invention of such formation, a plurality of faces in the image are divided into a plurality of set, and decision is used to determine become the selection set of the face of process object from a plurality of set.Then, final decision becomes the face of process object the face that comprised in the selection that determined set of decision unit.
Therefore, in image, exist under the situation of set of a plurality of faces, be not according to comprised in a plurality of set have the face and decide the face that becomes process object, but only decide the face that becomes process object according to the face that is comprised in any one set wherein.Therefore, only can carry out with by the corresponding control of face that is comprised in the set decision selection that the unit the determined set.For example, the collective that a plurality of people are arranged in image is separated and under the situation about existing, can carries out specific focusing and/or exposure control to any one collective, rather than the processing that these collectives are averaged.And, for example in image, exist under collective and isolated people's the situation, can not comprise this isolated people and only especially at collective, perhaps only especially at isolated people's focusing and/or exposure control.
The 6th mode of the present invention can constitute and also comprise display unit, and this display unit is the demonstration of being come by decision people's face that the unit determined and the difference of other face.Display unit is distinguished the demonstration of coming by for example display box around the people's face that is determined with other face.In addition, display unit showing the frame that has different colors, thickness or shape with frame shown around other face around the people's face that is determined, is distinguished come demonstration with other face by for example.In addition, display unit for example by the people's face that is determined being implemented the Flame Image Process different with other face and showing, is distinguished the demonstration of coming with other face.
(the 7th mode)
The 7th mode of the present invention is a kind of object determination device, comprises detecting unit and decision unit.Wherein, detecting unit adopts and the identical structure of the 6th mode of the present invention.
Decision unit in the 7th mode of the present invention is a benchmark the face of the people's face decision that is positioned at its center for the object that becomes focusing or exposure control with the number of the face that detected when being detected a plurality of people's face by detecting unit.This decision unit is counted the number of detected face, decides face according to the intermediate value of the number of being counted.For example, this decision unit sorts each face according to its x coordinate, is the above-mentioned face decision suitable with order intermediate value the face that becomes object.This decision unit also can decide the face that becomes object according to y coordinate rather than x coordinate.And, this decision unit can constitute according to x coordinate and y coordinate and select face respectively, under the situation of having selected same face, is this face decision the face that becomes object, under the situation of having selected different faces, benchmark according to the rules any one face decision wherein for becoming the face of object.
According to the 7th mode of the present invention, can obtain similarly and constitute the identical effect of the 6th mode of the present invention that decides the face that becomes object according to the center of the position of a plurality of faces.And,, there is no need to obtain the center of the position of a plurality of faces according to the 7th mode of the present invention.That is,, there is no need to carry out geometrical calculation as long as it is just much of that to carry out the counting and the data sorting of face.Therefore, under the situation of using general signal conditioning package, compare with the 6th mode of the present invention and can carry out processing apace.
And the 7th mode of the present invention can be the same with the 6th mode of the present invention, constitutes also to comprise taxon and set decision unit.In this case, the decision unit in the 7th mode of the present invention is according to people's face of selecting to be comprised in the set, and decision becomes the face of object.
(all directions formula)
All directions of the present invention formula is a kind of object determination device, comprises detecting unit, taxon, tentative decision unit and final decision unit.Wherein, detecting unit adopts and the identical structure of the 6th mode of the present invention with taxon.
Tentative decision unit in the formula of all directions of the present invention is in by each set in a plurality of set that taxon generated, and tentative decision becomes the face of the object of focusing or exposure control.In other words, tentative decision unit is at each set in a plurality of set, and the people's face that is comprised in this set, decision becomes the face of the object of focusing and/or becomes the face of the object of the control that exposes when implementing shooting.At this moment, fixing tentatively the decision unit decides the face that becomes object all can according to which kind of benchmark.For example, tentative decision unit can by with the identical processing in decision unit shown in the 6th mode of the present invention or the 7th mode, decide the face that becomes object.In addition, for example, tentative decision unit can determine the stipulated standard of unit according to the face of the 6th mode of the present invention, decides the face that becomes object.
Of the present invention the from all directions the final decision unit in the formula from become the face of the object of focusing or exposure control by final decision the tentative decision face that the unit determined.For example, the final decision unit can by with the identical processing in decision unit shown in the 6th mode of the present invention or the 7th mode, from each set, become the face of object by final decision in the tentative decision face that the unit determined.In addition, for example, the final decision unit can determine the stipulated standard of unit according to the face of the 6th mode of the present invention, and by in the tentative decision face that the unit determined, final decision becomes the face of object from each set.
According to all directions of the present invention formula, can obtain and constitute the 6th mode of the present invention or the identical effect of the 7th mode that also comprise taxon and set decision unit.And,, under keeping, can when decision becomes the face of object, decide this face according to various benchmark because of the state that comprises the effect that taxon and set decision unit obtain according to all directions of the present invention formula.
(the 9th mode)
The 9th mode of the present invention is a kind of object determination device, comprises detecting unit, decision unit, district and decision unit.
Detect people's face in each district in a plurality of districts that the detecting unit in the 9th mode of the present invention obtains in that the image of being imported is cut apart.Can be predetermined the image that to be imported and be divided into which type of district, also can dynamically determine.
Decision unit in district is according to the testing result of detecting unit, and decision is used to determine to become the face of the object of focusing when implementing shooting and/or becomes the district (district that comprises the face of the object that becomes focusing or exposure control: be called " selecting the district ") of face of the object of the control that exposes.For example, can decide the district according to the benchmark identical with set decision unit in the 6th mode of the present invention.
Decision becomes the face of object the face that decision unit in the 9th mode of the present invention is comprised in selecting the district.At this moment, which kind of benchmark this decision unit can decide the face that becomes object all can according to.For example, this decision unit can by with the identical processing in decision unit shown in the 6th mode of the present invention or the 7th mode, decide the face that becomes object.In addition, for example, this decision unit can determine the stipulated standard of unit according to the face of the 6th mode of the present invention, and decision becomes the face of object.
(the tenth mode)
The tenth mode of the present invention is a kind of object determination device, comprises detecting unit, judging unit, selected cell and decision unit.Wherein, detecting unit adopts and the identical structure of the 6th mode of the present invention.
When judging unit detects a plurality of face at detecting unit, judge the face of the maximum in detected people's face.Judging unit can be judged size according to the pixel count of the colour of skin in for example detected people's face.In addition, judging unit can according to for example detected when people's face the size of employed face rectangle judge the size of face.Judging unit constitutes according to other which kind of benchmark is judged that the size of face all can.In addition, it is proper maximum that maximum face there is no need, and can be the 2nd big face for example, also can be the 3rd big face.
Selected cell is selected the face of the maximum in the detected face and is had other face based on the size in the specialized range of the size of this face.
Decision unit in the tenth mode of the present invention determines the face that becomes the face of the object of focusing and/or become the object of the control that exposes when implementing shooting from by the selected face of selected cell.Like this, decision unit in the tenth mode of the present invention is not according to deciding the face that becomes object by the detected whole position of a plurality of faces of detecting unit, but the face of the maximum from detected face and having in other face based on the size in the specialized range of the size of this face, decision becomes the face of object.All can according to the face of which kind of benchmark decision becoming object in this decision unit.For example, this decision unit can by with the identical processing in decision unit shown in the 6th mode of the present invention or the 7th mode, decide the face that becomes object.In addition, for example, this decision unit can determine the stipulated standard of unit according to the face of the 6th mode of the present invention, and decision becomes the face of object.
According to the tenth mode of the present invention,, be selected to the face of focusing object only according to the face that in focusing, has size relatively to a certain degree with image.Therefore, can prevent to be the people of background that promptly the user does not have a mind to include as process object as people's face of subject being taken.Therefore, for example carry out under the situation of the processing identical with decision unit in the 6th mode of the present invention in this decision unit, improved the set of selecting to comprise the face that becomes the focusing object processing, obtain based on the processing at the center of the face that is comprised in the set and the precision of processing that is selected to the face of the object of focusing.
(the 11 mode)
The 11 mode of the present invention is a kind of object determination device, comprises detecting unit, taxon, set decision unit and decision unit.Wherein, detecting unit, taxon and set decision unit adopt the structure identical with the 6th mode.
Decision unit in the 11 mode of the present invention is from being become the face of object by decision the face that is comprised in the set decision set that the unit determined.At this moment, all can according to the face of which kind of benchmark decision becoming object in this decision unit.For example, this decision unit can by with the identical processing in decision unit shown in the 6th mode of the present invention or the 7th mode, decision becomes the face of object.In addition, for example, this decision unit can determine the stipulated standard of unit according to the face of the 6th mode of the present invention, and decision becomes the face of object.
According to the 11 mode of the present invention, can obtain and the identical effect of all directions formula of the present invention.
(the 12 mode)
The 12 mode of the present invention is a kind of object determination device, comprise detecting unit, characteristic quantity obtain the unit and the decision unit.Wherein, detecting unit adopts and the identical structure of the 6th mode of the present invention.
Characteristic quantity is obtained the unit at each of a plurality of people's faces detected in detecting unit, obtains the characteristic quantity of expression people face degree.Whether " people's face degree " can be according to being that the distance of the identification boundary line of people's face provides apart from being used to divide certain image for example.Obtained value was represented when people's face degree can use the face in detecting unit to detect processing.
The decision unit is according to characteristic quantity, and decision becomes the face of the object of focusing and/or becomes the face of the object of the control that exposes when implementing shooting.
(distortion of the 6th mode~the 12 mode)
The the 6th~12 mode of the present invention can constitute and comprise that also frame obtains the unit, and its image of frame of obtaining regulation from the sequential chart picture of being imported that comprises a plurality of frames is as the image of process object.Under situation about constituting like this, detecting unit detects people's face from the image of the frame of being obtained the obtained regulation in unit by frame.
In the 6th~12 mode of the present invention that constitutes like this,, can obtain and the identical effect/effect of the 6th~12 mode of the present invention for the sequential chart picture that comprises a plurality of frames.
(other)
First mode of the present invention~the 12 mode can realize by using the signal conditioning package executive routine.That is, the present invention can be defined as the performed processing in each unit in above-mentioned first mode~the 12 mode being used to making program that signal conditioning package carries out or the recording medium that records this program.
According to the present invention, when detecting subject, making a video recording there is no need to make a video recording has necessarily carried out the image of focusing to this subject as preparing.Therefore, in preparation shooting, be not used to obtain the focal point control of a plurality of images that carried out focusing etc., only use the obtained image of focus information according to the rules just can detect the main objects body apace.
And, according to the present invention, in preparation shooting, detect under a plurality of personages' the situation of face, according to the position of detected a plurality of faces and size etc., judge that automatically which face of reply focuses or expose control.
Description of drawings
Fig. 1 is the figure of the functional block of expression camera head.
Fig. 2 is that expression uses face to judge the figure of test example of the face of rectangle.
Fig. 3 is the figure that the expression face is judged the figure example of rectangle.
Fig. 4 is that expression uses face to judge the figure of test example of the face of rectangle.
Fig. 5 is the process flow diagram of the action example of expression camera head.
Fig. 6 is the process flow diagram of the action example of expression focusing object determination section.
Fig. 7 is the figure of the processing example of expression camera head.
Fig. 8 is the figure of the functional block of expression first embodiment.
Fig. 9 is used for figure that external polygon method is described.
Figure 10 is used for figure that gravity model appoach is described.
Figure 11 is used for the figure that the processing to the face determination section describes.
Figure 12 is the process flow diagram of the action example of expression second embodiment.
Figure 13 is the process flow diagram of a part of the action example of expression second embodiment.
Figure 14 is the process flow diagram of a part of the action example of expression second embodiment.
Figure 15 is the figure of the functional block of expression the 3rd embodiment.
Figure 16 is the figure of the processing example of presentation class portion.
Figure 17 is the figure of the processing example of the determination section in expression the 3rd embodiment.
Figure 18 is the process flow diagram of a part of the action example of expression the 3rd embodiment.
Figure 19 is the figure that is used for representing the effect of the 3rd embodiment.
Figure 20 is the figure of the functional block of expression the 4th embodiment.
Figure 21 is the process flow diagram of a part of the action example of expression the 4th embodiment.
Figure 22 is the figure of the functional block of expression the 5th embodiment.
Figure 23 is that expression is divided into the figure of the focusing in a plurality of districts with example images.
Figure 24 is the figure of the processing example of the determination section in expression the 5th embodiment.
Figure 25 is the process flow diagram of a part of the action example of expression the 5th embodiment.
Figure 26 is the figure of the functional block of expression the 6th embodiment.
Figure 27 is the figure of the processing example of the maximum face selection portion of expression.
Figure 28 is the figure of the processing example of expression candidate selection portion.
Figure 29 is the process flow diagram of a part of the action example of expression the 6th embodiment.
Symbol description
1: camera head; 2: image pickup part; 3: image storage part; 4: focusing object determination section; 5: the face test section; 6: determination section; 7: range finding portion; 8: face is judged rectangle; 9: the first rectangles; 10: the second rectangles; 11: input part; 12a, 12b, 12c, 12d, 12e: camera head; 13: input part; 14: image pickup part; 15: image storage part; 16a, 16b, 16c, 16d, 16e: focusing object determination section; 17a, 17d: face test section; 18a, 18b, 18c, 18d, 18e: determination section; 19a, 19b, 19d, 19e: center determination section; 20a, 20b, 20d, 20e: face determination section; 21: range finding portion; 22: division; 23: the set determination section; 24: district's determination section; 25: maximum face selection portion; 26: candidate's selection portion.
Embodiment
[first embodiment]
Below, use figure describes the camera head 1 of first embodiment.In addition, only be illustration at the following explanation of camera head 1, its structure is not limited to the following description.
(system architecture)
Camera head 1 is digital camera, digital camera or has digital camera and the signal conditioning package of the camera function that digital camera is such.Camera head 1 has CPU (central arithmetic processing apparatus), main storage means (RAM (the Read Only Memory that connects by bus on hardware, ROM (read-only memory))), auxilary unit and each device (pick-up lens, mechanical mechanism, CCD (Charge-Coupled Device, charge-coupled device (CCD)), operating portion, motor etc.) of being used for moving etc. as digital camera or digital camera.Auxilary unit uses Nonvolatile memory devices and constitutes.Here said Nonvolatile memory devices is meant so-called ROM (Read-Only Memory, ROM (read-only memory): comprise EPROM (Erasable Programmable Read-Only Memory, Erasable Programmable Read Only Memory EPROM), EEPROM (Electrically Erasable Programmable Read-Only Memory, Electrically Erasable Read Only Memory), mask model ROM etc.), FRAM (Ferroelectric RAM, ferroelectric RAM), hard disk etc.
Fig. 1 is the figure of the functional block of expression camera head 1.Camera head 1 is carried out by the various programs (OS, application program etc.) in the auxilary unit of being stored in are loaded in the main storage means and by CPU, thereby brings into play function as the device that comprises image pickup part 2, image storage part 3, focusing object determination section 4, range finding portion 7 and input part 11 etc.Focusing object determination section 4 is realized by using CPU to carry out focusing object determination procedure.In addition, focusing object determination section 4 also can be used as specialized hardware and realizes.
Below, use Fig. 1 that camera head 1 each included function portion is described.
((image pickup part))
Image pickup part 2 uses pick-up lens, mechanical mechanism, CCD, motor etc. to constitute the device with auto-focus function.Pick-up lens comprises the zoom lens that for example is used to realize zoom function, it is first-class to be used for focus lamp that any subject is focused.Mechanical mechanism comprises mechanical shutter, aperture, light filter etc.Motor comprises zoom lens motor, focus motor, fast door motor etc.
Image pickup part 2 has does not make illustrated storage part, and the focus information of regulation is stored in this storage part.The focus information of regulation is a focus information that is predetermined.Under the situation of the power connection of camera head 1, if do not have the input of focus information from range finding portion 7, then the amasthenic lens of image pickup part 2 is controlled so as to the state that becomes based on the focus information of this regulation.On the other hand, if from range finding portion 7 input of focus information is arranged, then the amasthenic lens of image pickup part 2 is controlled so as to the state that becomes based on the focus information of being imported.
Image pickup part 2 uses CCD that the image transitions of the subject of imaging by the pick-up lens that comprises controlled amasthenic lens is as described above become electric signal, thereby makes a video recording.
Image pickup part 2 can constitute when the user has carried out input by input part 11, just make a video recording 1 or many still images.And image pickup part 2 also can constitute the input of being undertaken by input part 11 according to the user, the dynamic image (sequential chart picture) that shooting is made of a plurality of frames.And image pickup part 2 can constitute the image of obtaining a frame from a plurality of frames under the situation that constitutes the shooting dynamic image, gives focusing object determination section 4 image of this frame as the image transfer of process object.
((image storage part))
Read-write recording mediums such as the so-called RAM of image storage part 3 uses wait and constitute.Image storage part 3 also can use with respect to camera head 1 removable recording medium and constitute.The data that image storage part 3 is stored the image of being made a video recording by image pickup part 2.
((focusing object determination section))
Focusing object determination section 4 subject that decision is focused from the subject of being made a video recording by image pickup part 2 promptly becomes the subject of the object of focusing.At this moment, focusing object determination section 4 uses focus information according to the rules to be handled for the image of using and making a video recording of focusing by image pickup part 2.The image should to use and make a video recording in order to focus is input in the focusing object determination section 4 by image pickup part 2.
Focusing object determination section 4 comprises: face test section 5, and it detects the personage's who becomes subject face from the image of being imported; And determination section 6, it is according to the subject of being focused by the 5 detected face decisions of face test section.Below, each the function portion that constitutes focusing object determination section 4 is described.
<face test section 〉
Face test section 5 detects the personage's who becomes subject face according to the statistic (specifically deep or light value) of a plurality of subregions from the image of being imported.Below, the processing to by face test section 5 face performed, that detect the personage time is elaborated.
Fig. 2 is the figure of expression based on the detection method of the personage's of deep or light value face.In this detection method, specific rectangle (hereinafter referred to as " face judgement rectangle ") 8 is moved in becoming the image of process object, judge at each face that moves the destination and judge whether (hereinafter referred to as " region-of-interest ") in the rectangle 8 includes personage's face.At this moment, by changing the angle that face is judged the ratio of the size of rectangle 8 and the size of the image that becomes process object, face judgement rectangle 8 and become the image of process object, can detect the face of all size or angle.
In the rectangle of face judgement rectangle 8, comprise a plurality of other rectangles (hereinafter referred to as " first rectangle ", " second rectangle ") 9,10.First rectangle 9 and second rectangle 10 are configured in the fixed position in the face judgement rectangle 8 respectively.In judging rectangle 8, a face disposes first rectangle 9 and second rectangle 10 more than 1 respectively.
In the detection of face is handled, calculate in each region-of-interest the deep or light value of each zone that is surrounded by first rectangle 9 and second rectangle 10 (below be called " first characteristic area ", " second characteristic area ").The deep or light difference of first characteristic area and second characteristic area is for example represented by the mean value or the aggregate value of the pixel value in each zone.
Then, calculate the residual quantity of the deep or light value Lb of the deep or light value La of first characteristic area and second characteristic area.Then, whether greater than pre-set threshold α, judge the face that in region-of-interest, whether includes the personage according to this residual quantity value.
Fig. 3 is the figure that the expression face is judged the figure example of rectangle 8.Fig. 4 is that expression uses the face of a plurality of figures to judge the figure of the face determination processing of rectangle 8.In actual treatment, as shown in Figure 3, judge rectangle 8 for face and prepare to have a plurality of figures.In each figure, the quantity and the position of first rectangle 9 and second rectangle 10 have nothing in common with each other.
Then, carry out judgement at the face judgement rectangle 8 of each figure by calculating deep or light difference.And, according to each result of determination, calculate score at each region-of-interest, according to the score of being calculated, the face that whether includes the personage in this region-of-interest is carried out final judgement.Face corresponding that whether includes the personage in the score of being calculated and the region-of-interest be by the more sample image of for example usage quantity (comprise the personage face image and do not comprise the image of personage's face) study handle and obtain in advance.
Face test section 5 is detecting under the situation of face, has promptly carried out including in certain region-of-interest under personage's the situation of final decision of face, and the image information relevant with this face outputed to determination section 6.Specifically, face test section 5 output packets contain the position and the size of the region-of-interest of detected face.For example, the centre coordinate of face test section 5 these region-of-interests of output and area (pixel count in the zone).
And, face test section 5 can constitute as required to detected face towards judging and exporting.In addition, face test section 5 judge face towards situation under, judge by for example using non-patent literature 1,2 technology of being put down in writing.Specifically, face test section 5 for example also uses a left side to judge that towards the face of the figure of usefulness the face of rectangle 8 and right figure towards usefulness judges that rectangle 8 carries out face and detect.And face test section 5 is for example using a left side to judge that towards the face of the figure of usefulness rectangle 8 detects under the situation of face, be judged to be detected face and be a left side towards.Equally, face test section 5 judges that at the face that uses right figure towards usefulness rectangle 8 detects under the situation of face, be judged to be detected face and be right towards.For the front too.The orientation judging of face can be used any technology and implement and be not limited to above-mentioned technology.
<determination section 〉
Determination section 6 determines the subject of focusing according to the image information relevant with face from 5 inputs of face test section.At this moment, roughly come to divide, determination section 6 carries out two processing.First to handle be from by the processing of selecting a face the detected a plurality of faces of face test section 5.Second processing is the face of selecting according to by first treatment of selected, the processing of the subject that decision is focused.In addition, in face test section 5, only detect under the situation of a face, omit first and handle.And, in face test section 5, do not detect under the situation of face, omit first and handle and second processing.
<<the first handles〉〉
At first, first processing is described.In first handled, determination section 6 will use the image information of which face when being chosen in the subject that decision is focused in second processing.In first handles, three kinds of implementation methods that are called as area relative method, score relative method, telemetry are arranged.In determination section 6, can adopt any one method wherein.Below, each implementation method is described.
<<<area relative method〉〉 〉
In the area relative method, the area of 6 pairs of detected a plurality of faces in face test section 5 of determination section compares, and selecting the area as image is maximum face.In the area relative method, determination section 6 is in order to detect apart from the nearest face of camera head 1, and selecting the area as image be the face of maximum.That is, in the area relative method, determination section 6 compares the comparison of the distance that is implemented into subject by the size to the face image.
<<<score relative method〉〉 〉
In the score relative method, determination section 6 is obtained score respectively at a plurality of faces detected in face test section 5.Determination section 6 obtains score according to the face density around position in image of the size (area) towards, face of face, face, the face etc.Determination section 6 can also can obtain score at part item at above-mentioned all four acquisition scores.
Then, determination section 6 selects to have obtained the face of top score.About face towards, for example face towards towards the front, then score is set highly more.About the size of face, for example the size of face is big more, and then score is set highly more.About the position of face in image, for example near more apart from image central authorities, then score is set highly more.About face density, for example face density is high more, and then score is set highly more.Yet, in every (towards, size, position, face density), be which kind of setting state high score, can design by the user variablely, also can design arbitrarily by the deviser.
<<<telemetry〉〉〉the meaning design.
<<<telemetry〉〉 〉
In telemetry, determination section 6 is obtained at the range finding result by the material object of each detected face of face test section 5.And determination section 6 is selected apart from the minimum face of the distance of camera head 1.
Determination section 6 can constitute the range finding result who obtains any distance measuring equipment.For example, can constitute in camera head 1 or focusing object determination section 4, to be equipped with and not make illustrated distance measuring equipment, obtain the range finding result of this distance measuring equipment by determination section 6.And, for example, under range finding portion 7 adopts situation to the structure of finding range to the distance of subject, can be used as range finding portion 7 replacement device of distance measuring equipment.In this case, determination section 6 obtains the range finding result (with reference to the dotted arrow of Fig. 1) of range finding portion 7.Under any situation, determination section 6 offers distance measuring equipment to the coordinate of the region-of-interest from the image information of face test section 5 inputs.And distance measuring equipment is according to the coordinate of this region-of-interest, to finding range to the distance of detected face.
<<the second handles〉〉
In second handled, determination section 6 determined the subject of focusing according to selected face.As the example of this subject, the position of the face that is allowed a choice (for example eye, nose, mouth, ear, volume etc.).At this moment, the position at the position of selected face can be detected again by face test section 5, also can obtain its position according to the image information of selected face.For example, determination section 6 is being the position at the position of selected face from relatively the be separated by location recognition of predetermined distance of the centre coordinate of selected face on prescribed direction.Then, determination section 6 is exported to range finding portion 7 to the coordinate (being two-dimensional coordinate) of the image of the subject of focusing.
And selected face itself can be the subject of focusing.In this case, determination section 6 is exported to range finding portion 7 to the position (being two-dimensional coordinate) of the face that is comprised in the image information of selected face in first handles.
((range finding portion))
Range finding portion 7 obtains the focus information that is used for subject focusing that the determination section 6 by focusing object determination section 4 is determined.At this moment, range finding portion 7 is according to the two-dimensional coordinate from determination section 6 inputs, the subject that is identified for focusing.
For example, range finding portion 7 constitutes by penetrate infrared ray to subject and waits the distance to subject is found range, thereby obtains focus information.In this case, range finding portion 7 according to from
((input part))
Input part 11 use shutter release buttons wait and constitute.When input part 11 detects user's the input of order,, each one of camera head 1 is carried out the order notice according to the order of being imported.For example, when input part 11 detects user's the input of self-focusing order, notify self-focusing order to image pickup part 2.And, for example, when input part 11 detects user's the input of shooting order, to the order of image pickup part 2 notice shootings.
Under the situation that input part 11 use shutter release buttons constitute, the automatic focus order is detected by input part 11 by for example this shutter release button partly being pressed.And the shooting order is detected by input part 11 by for example this shutter release button being pressed fully.
(action example)
Fig. 5 is the process flow diagram of the action example of expression camera head 1.Below, use Fig. 5 that the action example of camera head 1 is described.
When camera head 1 energized, image pickup part 2 is controlled to amasthenic lens the state that becomes based on the focus information of regulation.
When input part 11 detects shutter release button and partly pressed by the user (S01), give image pickup part 2 advisory of having imported the automatic focus order.
Image pickup part 2 judges whether to have imported focus information from range finding portion 7.Under the situation of input focus information not (S02-is not), image pickup part 2 under amasthenic lens is controlled so as to state based on the focus information of regulation, make a video recording (S03).Image pickup part 2 is exported to focusing object determination section 4 to the view data of being made a video recording.
Focusing object determination section 4 determines the subject (S04) of focusing according to the view data of being imported.Fig. 6 is the process flow diagram of the action example of expression focusing object determination section 4.Use Fig. 6 that the processing of S04 is described.
At first, face test section 5 detects personage's face (S10) from the image of being imported.Face test section 5 is exported to determination section 6 to the image information of detected face.
Determination section 6 investigation is by the number of the detected face of face test section 5.(S11:<1), be under 0 the situation, determination section 6 is exported to range finding portion 7 to the two-dimensional coordinate of picture central authorities under less than 1 situation when the number of face.(S11:>1), promptly detecting under the situation of a plurality of faces, determination section 6 is carried out first and is handled, and selects a face (S13) from detected a plurality of faces under greater than 1 situation at the number of face.
Be (S11:=1) under one the situation at detected face, perhaps after the processing of S13, determination section 6 decides the subject (S14) of focusing according to a face.Here a said face is to be this unique face that is detected under one the situation at the number of detected face, is selecteed face under the situation after S13 handles.And determination section 6 is obtained the two-dimensional coordinate of the subject that is determined, this two-dimensional coordinate is exported to range finding portion 7 (S15).
Get back to the explanation of using Fig. 5.After the processing of S04, range finding portion 7 obtains the focus information (S05) at the subject that is determined according to the two-dimensional coordinate of output from the determination section 6 of focusing object determination section 4.Range finding portion 7 exports to image pickup part 2 to obtained focus information.After this is handled, carry out S02 and processing afterwards once more.
In S02,2 having under the situation of input of focus information (S02-is) from range finding portion 7 to image pickup part, image pickup part 2 is controlled amasthenic lens (S06) according to the focus information of being imported.That is, image pickup part 2 control amasthenic lens are so that the subject focusing to being determined by focusing object determination section 4.
After the control of amasthenic lens, image pickup part 2 make a video recording (S07).By this shooting, the image that shooting has been carried out focusing to the subject that is determined by focusing object determination section 4.And, the data (S08) that image storage part 3 is stored the image of being made a video recording by image pickup part 2.
(effect/effect)
Fig. 7 is the figure of the example of the expression image of being made a video recording by camera head 1.In camera head 1, detect face as the main objects body (face that fences up by the thick lines rectangle), rather than as a setting and the personage's who is made a video recording face, so that the mode of this main objects body focusing is implemented shooting.
In camera head 1, the face test section 5 of focusing object determination section 4 detects personage's face as the subject that becomes the focusing object.At this moment, face test section 5 carries out the face detection according to the deep or light difference of first characteristic area in the region-of-interest and second characteristic area.During deep or light value in obtaining certain zone, even under the situation that the subject of being made a video recording in not to this zone is focused, the value under the situation that also can obtain to approach to carry out to focus.Therefore, in camera head 1, even under the situation of the image of personage's face of having made a video recording out of focus, also can detect this face.Therefore, in camera head 1, when shooting detects personage's face during employed image, there is no need to carry out focal point control to subject.Therefore, when this shooting is handled, in camera head 1, there is no need to carry out focal point control again, can carry out shooting apace at the amasthenic lens of the image pickup part 2 that is controlled as the regulation focus.
And focusing object determination section 4 is only at an image, promptly is controlled as the regulation focus and the image of making a video recording carries out face and detects.Therefore, in camera head 1, there is no need that a plurality of images are carried out face and detect, can detect face apace.
And, in camera head 1, detecting under the situation of a plurality of faces, focusing object determination section 4 is selected any one face, according to the subject of this face decision becoming focusing object.Therefore, according to camera head 1, the user there is no need specially to select a face a plurality of faces in being incorporated in frame.
(variation)
Focusing object determination section 4 can independently constitute the external device (ED) of camera head 1, and is not included in the camera head 1.And the focusing object determination section 4 that constitutes external device (ED) like this can constitute and be equipped with the illustrated range finding portion that do not do that has with range finding portion 7 identical functions in the device of self.
In addition, camera head 1 is obtained focus information by using 7 pairs of distances to subject of range finding portion to find range, and uses range finding method in addition to obtain focus information yet also can constitute.
In addition, camera head 1 also can constitute the subject that becomes the object of the control that exposes according to detected face decision when implementing shooting, and control thus exposes.In this case, determination section 6 decisions become the subject of the object of the control that exposes.And in this case, camera head 1 also has photometric measurer, and this photometric measurer is obtained the exposure control information according to the subject that is determined by determination section 6.Photometric measurer is calculated exposure (exposure control information) by the subject that is determined by determination section 6 is carried out photometry as the exposure area.Then, photometric measurer is delivered to image pickup part 2 to the exposure control information of being calculated.Then, image pickup part 2 waits the exposure control of implementing based on this exposure control information by regulating aperture, makes a video recording.
In addition, face test section 5 also can constitute the subject detection with colour of skin is face.By such formation, although can not detect personage's face exactly, handle the needed time yet can cut down to detect, also can cut down the needed time of installation and the cost of face test section 5.
In addition, image pickup part 2 also can be used as the Manual focusing device and realizes, rather than auto-focus function.In this case, for example can constitute, when becoming the state that is suitable for from the focus information of range finding portion 7 inputs, the situation of having carried out focusing be reported to the user when having operated amasthenic lens by the user.
In addition, camera head 1 also can constitute the argentalium camera on the film of making a video recording, rather than uses the image pickup part 2 with CCD finally to implement shooting.In this case, camera head 1 has the image pickup part 2 that possesses CCD and as the image pickup part of argentalium camera.And in this case, image storage part 3 is not necessary for camera head 1.Image pickup part 2 is implemented shooting for the image of preparation shooting being input in the focusing object determination section 4.And, in as the image pickup part of argentalium camera, implement the adjusting based on the amasthenic lens of focus information of range finding portion 7.Yet, image pickup part 2 and can constitute the part of first-class its structure of shared for example focus lamp as the image pickup part of argentalium camera.
[summary of second~the 6th embodiment]
Below, use figure describes the camera head 12 (12a, 12b, 12c, 12d, 12e) with focusing object determination section 16 (16a, 16b, 16c, 16d, 16e).In this explanation, character image is part or all the image of image that comprises personage's face at least.Therefore, character image can comprise the image of personage's integral body, also can comprise personage's only face or above the waist image only.And, can comprise image at a plurality of personages.And background can comprise any figures such as landscape beyond the personage (background: also comprise as subject and the object of being paid close attention to) and pattern.
In addition, only be illustration at the following explanation of focusing object determination section 16 and camera head 12, its structure is not limited to following explanation.
[second embodiment]
(system architecture)
At first, the camera head 12a with focusing object determination section 16a is described.Camera head 12a has CPU (central arithmetic processing apparatus), main storage means (RAM), the auxilary unit that connects by bus and each device (pick-up lens, mechanical mechanism, CCD, operating portion, motor etc.) of being used for moving as digital camera and digital camera etc. on hardware.Auxilary unit uses Nonvolatile memory devices to constitute.Here said Nonvolatile memory devices is meant so-called ROM, FRAM, hard disk etc.
Fig. 8 is the figure of the functional block of expression camera head 12a.Camera head 12a is by being loaded into the various programs (OS, application program etc.) in the auxilary unit of being stored in the main storage means and using CPU to carry out, thereby brings into play function as the device that comprises input part 13, image pickup part 14, image storage part 15, focusing object determination section 16a and range finding portion 21 etc.Focusing object determination section 16a realizes by using CPU to carry out focusing object determination procedure.In addition, focusing object determination section 16a also can be used as specialized hardware and realizes.
Below, use Fig. 8 that each the function portion that is comprised in the camera head 12a is described.
<input part 〉
Input part 13 use shutter release buttons wait and constitute.When input part 13 detects user's the input of order,, each one of camera head 12a is carried out the order notice according to the order of being imported.For example, when input part 13 detects user's the input of automatic focus order, to image pickup part 14 notice automatic focus orders.And, for example, when input part 13 detects user's the input of shooting order, to image pickup part 14 notice shooting orders.
Under the situation that input part 13 use shutter release buttons constitute, the automatic focus order is by for example this shutter release button partly being pressed, being detected by input part 13.And the shooting order is by for example this shutter release button being pressed fully, being detected by input part 13.
<image pickup part 〉
Image pickup part 14 uses pick-up lens, mechanical mechanism, CCD, motor etc., constitutes the device with auto-focus function.Pick-up lens comprises the zoom lens that for example is used to realize zoom function, it is first-class to be used for focus lamp that any subject is focused.Mechanical mechanism comprises mechanical shutter, aperture, light filter etc.Motor comprises zoom lens motor, focus motor, fast door motor etc.
Image pickup part 14 has does not make illustrated storage part, and the focus information of regulation is stored in this storage part.The focus information of regulation is a focus information that is predetermined.Under the situation of the power connection of camera head 12a, if do not have the input of focus information from range finding portion 21, then the amasthenic lens of image pickup part 14 is controlled so as to the state that becomes based on the focus information of this regulation.On the other hand, if from range finding portion 21 input of focus information is arranged, then the amasthenic lens of image pickup part 14 is controlled so as to the state that becomes based on the focus information of being imported.
Image pickup part 14 becomes electric signal to the image transitions of the subject of imaging via the pick-up lens that includes controlled amasthenic lens as described above by using CCD, makes a video recording.
Image pickup part 14 can constitute when the user has carried out input by input part 13, just make a video recording 1 or many still images.In addition, image pickup part 14 also can constitute according to the user's input by input part 13, the dynamic image (sequential chart picture) that shooting is made of a plurality of frames.In addition, image pickup part 14 can constitute the image of obtaining a frame from a plurality of frames under the situation that constitutes the shooting dynamic image, gives focusing object determination section 16a the image of this frame as the image transfer of process object.
<image storage part 〉
Read-write recording mediums such as the so-called RAM of image storage part 15 uses wait and constitute.Image storage part 15 also can use with respect to the removable recording medium of camera head 12a and constitute.The data that image storage part 15 is stored the image of being made a video recording by image pickup part 14.
<focusing object determination section 〉
Focusing object determination section 16a determine the subject of focusing from the subject of being made a video recording by image pickup part 14, promptly become the face of the object of focusing.At this moment, focusing object determination section 16a uses by image pickup part 14 and handles for the image of using and making a video recording of focusing.The image should to use and make a video recording in order to focus inputs to focusing object determination section 16a by image pickup part 14.
Focusing object determination section 16a comprises face test section 17a and determination section 18a.Below, each the function portion that constitutes focusing object determination section 16a is described.
<face test section 〉
Face test section 17a comprises the coordinate in zone (hereinafter referred to as " face zone ") of personage's face from focusing with acquisition the image.According to the coordinate of this face rectangle, determine position and the size of focusing with the face zone of the people in the image.
The detection in face zone can realize by using existing any technology.For example, can obtain the coordinate of face rectangle by the template matches of using the benchmark template corresponding with the profile of face integral body.In addition, also can obtain the coordinate of face rectangle by template matches based on the inscape (eye, nose, ear etc.) of face.In addition, also can handle the summit of detecting hair, obtain the coordinate of face rectangle according to this summit by chroma key.In addition, also can obtain the coordinate of face rectangle according to the detection of area of skin color.
<determination section 〉
Determination section 18a is from being become the face of focusing object by decision the detected face of face test section 17a.The coordinate (coordinate of for example representing the central authorities of face rectangle) of the face rectangle that determination section 18a handle is relevant with the face that is determined passes to range finding portion 21.Determination section 18a comprises center determination section 19a and face determination section 20a.Below, each the function portion that constitutes determination section 18a is described.
<center determination section 〉
Center determination section 19a obtains the center of the position of detected a plurality of faces being under a plurality of situation by the detected face of face test section 17a.Here, suppose the position of central point (hereinafter referred to as " face point ") the expression face of face rectangle.Be that center determination section 19a obtains the mid point of two face points under two the situation at detected face.And under detected face was situation more than three, center determination section 19a used following any method to obtain the center.
<<external polygon method〉〉
In external polygon method, center determination section 19a obtains and the center of the external polygonal center of a plurality of face points as the position of a plurality of faces.No matter used polygonal number of vertex is what are individual all passable, also can be predetermined.Here, be that 4 situation describes to number of vertex.
Fig. 9 is used for center determination section 19a is obtained the figure that the processing at center describes according to external polygon method.Center determination section 19a at the x coordinate and the y coordinate of a plurality of face points, judges maximal value and minimum value respectively.Center determination section 19a generates by the straight line parallel with the y axle and the rectangle that straight line generated parallel with the x axle (in orthogonal coordinate system, constitute the rectangle on summit by the intersection point of each bar line of maximal value parallel with the x axle and each the coordinate composition by the face point coordinate respectively and minimum value) with the y axle, the maximal value of getting the x coordinate and the face point of minimum value should be straight line parallel comprised, the maximal value of getting the y coordinate and the face point of minimum value should be straight line parallel comprised with the x axle with the y axle.And center determination section 19a obtains the coordinate of intersection of diagonal of the rectangle that is generated as centre coordinate.
<<gravity model appoach〉〉
Figure 10 is used for center determination section 19a is obtained the figure that the processing at center describes according to gravity model appoach.In gravity model appoach, center determination section 19a obtains the center of gravity of a plurality of face points.Specifically, center determination section 19a by with the position vector of each face point and obtain the position vector of center of gravity divided by the number of face point, obtain barycentric coordinates according to this position vector.Then, determination section 19a in center gets obtained barycentric coordinates and makes centre coordinate.Figure 10 shows at 6 face points, obtains respectively apart from the position vector of certain reference point (point that for example utilizes the coordinate components minimum value separately of each face point to represent), obtains the situation of barycentric coordinates (Xg, Yg).
<face determination section 〉
Face determination section 20a basis is by the obtained centre coordinate of center determination section 19a, and which face decision can focus according to.Figure 11 is used for the figure that the processing to face determination section 20a describes.Face determination section 20a benchmark according to the rules selects face from being arranged in apart from the face of center gauge set a distance.At this moment, face determination section 20a according to the position of the size of face, face, face towards etc. in which benchmark decide face all can.For example, face determination section 20a is being arranged in apart from selected the nearest face in central point (Figure 11 is by symbol " * " diagram) apart from image (frame) (the face point that is fenced up by the thick lines circle) by the face of the obtained center gauge set a distance of center determination section 19a.
In example shown in Figure 11,, measure distance L 1 and L2 with the central point zero of image (frame) at being two face point F1 and the F2 that exists in the round C1 of radius at center F0 predetermined distance R with distance face point.Then, because L1<L2, thereby the face point (face of promptly focus object) of face point F1 decision for the focusing object.
Face determination section 20a passes to range finding portion 21 to the coordinate of selected face point.
<range finding portion 〉
Range finding portion 21 obtains the focus information that is used for face focusing that the face determination section 20a by focusing object determination section 16a is determined.At this moment, range finding portion 21 is according to the two-dimensional coordinate (face point coordinate) from face determination section 20a input, the subject that is identified for focusing.The range finding of being implemented by range finding portion 21 can application examples as penetrating the mode (active mode) that infrared ray is implemented by material object to face, also can use the mode (for example passive mode) beyond this active mode.
For example, range finding portion 21 constitutes by penetrate infrared ray etc. to subject the distance to subject is found range, thereby obtains focus information.In this case, the two-dimensional coordinate that range finding portion 21 bases are imported from face determination section 20a, the direction of decision ejaculation infrared ray etc.
When range finding portion 21 obtains focus information, obtained focus information is passed to image pickup part 14.
(action example)
Figure 12 is the process flow diagram of the action example of expression camera head 12a.Below, use Figure 12 that the action example of camera head 12a is described.
When camera head 12a energized, image pickup part 14 is controlled to amasthenic lens the state that becomes based on the focus information of regulation.
When input part 13 detects shutter release button and partly pressed by the user (S16), give image pickup part 14 advisory of having imported the automatic focus order.
Image pickup part 14 judges whether imported focus information from range finding portion 21.Under the situation of input focus information not (S17-is not), image pickup part 14 shooting focusings are with images (S18).Image pickup part 14 is given focusing object determination section 16a the focusing of being made a video recording with image transfer.
Focusing object determination section 16a is according to the focusing of being imported image, and decision becomes the face (S19) of the object of focusing.Figure 13 is the process flow diagram of the action example (example of (S19) is handled in the decision of the face of focusing) of expression focusing object determination section 16a.Use Figure 13 that the processing of S19 is described.
At first, face test section 17a detects personage's face (S24) from the image of being imported.Face test section 17a passes to the image information of detected face the center determination section 19a of determination section 18a.
Determination section 19a investigation in center is by the number of the detected face of face test section 17a.(S25:<1), be under 0 the situation, center determination section 19a is the advisory that the does not have face determination section 20 of saveing somebody's face under less than 1 situation when the number of face.And face determination section 20a passes to range finding portion 21 (S26) to the two-dimensional coordinate of image central authorities.
(S25:>1), promptly detecting under the situation of a plurality of faces, determination section 18a carries out the processing (S27) of selecting a face from a plurality of faces under more than 1 situation at the number of face.Figure 14 is the process flow diagram that the example of the processing (S27) of selecting a face from a plurality of faces is shown.Use Figure 14 that the processing of S27 is described.
Center determination section 19a obtains the centre coordinate (S29) of the face point of detected face.Center determination section 19a is the obtained centre coordinate transmission determination section 20a that savees somebody's face.Then, face determination section 20a adopts and for example uses the illustrated method of Figure 11 to select a face (S30) according to the centre coordinate of being imported.
Get back to the explanation of using Figure 13.At face detected from image is that perhaps after the processing of S27, face determination section 20a passes to range finding portion 21 (S28) to the two-dimensional coordinate of the face point of a face under one the situation (S25:=1).Here a said face is to be this unique face that is detected under one the situation at the number of detected face, is selecteed face under the situation after the processing of S27.
Get back to the explanation of using Figure 12.After the processing of S19, range finding portion 21 obtains focus information (S20) according to the two-dimensional coordinate that the face determination section 20a from focusing object determination section 16a is transmitted.Range finding portion 21 passes to image pickup part 14 to obtained focus information.After this is handled, carry out S17 and processing afterwards once more.
In S17,14 having under the situation of input of focus information (S17-is) from range finding portion 21 to image pickup part, image pickup part 14 is according to the focus information control amasthenic lens of being imported (S21).That is, image pickup part 14 control amasthenic lens are so that the face focusing to being determined by focusing object determination section 16a.
After the control of amasthenic lens, image pickup part 14 make a video recording (S22).By this shooting, the image of having made a video recording the face that is determined by focusing object determination section 16a having been carried out focusing.Then, the data (S23) of the image storage part 15 storage image of being made a video recording by image pickup part 14.
(effect/effect)
According to the camera head 12a as second embodiment of the present invention, the people's face that exists in the detected image is the focusing object to any one decision in the detected face automatically.Therefore, the user there is no need to make the face that will become the focusing object to be positioned at the center of image (frame).In other words, the user there is no need the face that will focus is placed on the center of frame.
And, according to camera head 12a, in image, exist under the situation of a plurality of faces as second embodiment of the present invention, according to the center of the position of a plurality of faces that exist in the image, decision becomes the face of focusing object.Therefore, even in image, exist under the situation of a plurality of faces, also can determine to become a face of focusing object.And, can focus near the face the central authorities of a plurality of faces of being arranged in set.
(variation)
Face determination section 20a can constitute benchmark according to the rules, from be arranged in apart from the center gauge set a distance and select face towards the face in front.At this moment, face determination section 20a uses for example following described technology of known document to judge that whether face is towards the front.
H.Schneiderman,T.Kanade.“A?Statistical?Method?for?3D?Object?Detection?Applied?to?Faces?and?Cars.”IEEE?Conference?on?Computer?Vision?and?Pattern?Recognition,2000.
And, in the foregoing description to center determination section 19a the position of face as the Computer Vision and Pattern Recognition that obtains of point, 2000.
And, in the foregoing description center determination section 19a is illustrated the position of face as the structure obtained of point, yet center determination section 19a also can constitute the position of face is obtained as the zone.
And, also can constitute the face that determined by face determination section 20a by not making illustrated display part, display box around this face.By such formation, the user knows which face is decided to be the focusing object.And, can be designed under the dissatisfied situation that the face that is shown is like this focused of user, by using input part 13 input commands, can be by manual operation so that to other face focusing.
And face determination section 20a can constitute benchmark according to the rules, selects face from being arranged in apart from center gauge set a distance and the face more than the prescribed level wherein.
And camera head 12a also can constitute the face that is determined according to by determination section 18a, carries out exposure control.This point is also identical in second to the 6th embodiment described later.
And determination section 18a also can will be sorted according to its x coordinate or y coordinate by each detected face of face test section 17a, the face decision suitable with the order of the intermediate value of the number of detected face for becoming the face of object.
And determination section 18a also can constitute benchmark according to the rules from by determining face the detected face of face test section 17a.Stipulated standard for example can be based on the benchmark (benchmark of the face of for example decision maximum of the size of face, the benchmark of the face of size in the middle of the decision, the benchmark of the face that decision is minimum), can be based on the benchmark benchmark of the nearest face of image central authorities (for example apart from) of the position of the face in the image, can be based on face towards the benchmark benchmark of the face in front (for example towards), can be based on the benchmark (amount that for example determines expression to be similar to the degree of face is the benchmark of the face of maximum) that is similar to face, the benchmark that can be based on the sex of inferring according to face (for example determines to be speculated as the benchmark of the male sex's face, decision is speculated as the benchmark of women's face), (decision is speculated as the benchmark of face in minimus age can be based on the benchmark in the age of inferring according to face, decision is by the benchmark of the face in age in the middle of being speculated as), be that other which kind of benchmark all can.And stipulated standard also can be the benchmark that said reference is suitably carried out a plurality of combinations.
And in the above description, prerequisite is that image pickup part 14 constitutes a subject is focused.Therefore, focusing object determination section 16a passes to range finding portion 21 to the coordinate of a face.Right is the example of this image pickup part 14, has the spy to open the described device of flat 11-295826.In this case, the face determination section 20a of determination section 5a benchmark according to the rules gives priority to face, selects the face of defined amount according to this priority.For example, face determination section 20a can constitute according to the face near to select progressively defined amount far away apart from the center that is determined by center determination section 19a.Number at selected face does not reach under the situation of stated number, uses other which kind of method to obtain the coordinate that passes to range finding portion 21 and all can.
[the 3rd embodiment]
Below, the camera head 12b with focusing object determination section 16b is described.Below, at camera head 12b and focusing object determination section 16b, illustrate with second embodiment in camera head 12a and the difference of the object determination section 16a that focuses.
(system architecture)
Figure 15 is the figure that the functional block of camera head 12b is shown.Replace focusing object determination section 16a, camera head 12b has focusing object determination section 16b, and this point is different with camera head 12a in second embodiment.Therefore, below, explanation omitted to input part 13, image pickup part 14, image storage part 15 and the range finding portion 21 of camera head 12b.
<focusing object determination section 〉
Focusing object determination section 16b comprises face test section 17a, division 22, set determination section 23 and determination section 18b.Below, each the function portion that constitutes focusing object determination section 16b is described.Yet, since focusing object determination section 16b included face test section 17a have with second embodiment in the included identical structure of face test section 17a of focusing object determination section 16a, thereby omit its explanation.
<division 〉
Division 22 is categorized as a plurality of set to the face point of detected face in face test section 17a.This classification can be used any methods such as nearest neighbor algorithm.Division 22 is carried out classification according to the position of each face point.The classification results of division 22 is each face point to be categorized into the close face point in position that makes in the image be comprised in the identity set.
Figure 16 is the figure that the processing example of division 22 is shown.As shown in figure 16, under the situation of the collective that has a plurality of face points, each collective is categorized as a set.
<set determination section 〉
Being used in a plurality of set of set determination section 23 decision classification in division 22 determines to become the set (promptly comprising the set of the face that becomes the focusing object: hereinafter referred to as " selecting set ") of the face of focusing object.That is, the face point coordinate that passes to range finding portion 21 is the face point coordinate of the face that comprised in the selection set that is determined by set determination section 23.
Set determination section 23 decides selection to gather according to the number of the face that comprised in each set, which kind of benchmark such as characteristic quantity relevant with the face that comprised in each set all can.For example, set determination section 23 is the selection set to the set decision that comprises maximum faces in each set.And for example, relevant set decision characteristic quantity and maximum of the face that is comprised in 23 of determination sections of set and each set is gathered for selecting.The characteristic quantity relevant with face can wait according to the number towards, face of the size of for example face, face and obtain.And the characteristic quantity relevant with face can be the value of expression people face degree.Whether people's face degree for example can be according to being that the distance of the identification boundary line of people's face provides apart from being used to divide certain image.People's face degree also can be used and detect when handling obtained value at the face of face test section 17a and represent.
Use the processing example of Figure 16 pair set determination section 23 to describe.Benchmark (be here in the set face that comprised count many benchmark) according to the rules, by set determination section 23 from by selecting upper left set to gather in a plurality of set that processing generated (being three set Figure 16) of division 22 as selecting.
<determination section 〉
Determination section 18b is not the face that is comprised according in the integral image of being made a video recording by image pickup part 14, but handle (being the superclass of the decision of face) to the choice set cooperation according to the face that is comprised in the selection set that is determined by set determination section 23, this point is different with determination section 18a.Determination section 18b carries out the processing identical with determination section 18a except above-mentioned point.
Figure 17 is the figure that the processing example of determination section 18b is shown.In the selection set by 23 decisions of set determination section, obtain the center of face point by the center determination section 19b of determination section 18b.Then, according to obtained center, by face determination section 20b, benchmark (being the near benchmark in central point apart from image (frame) here) is selected a face point (the face point that is fenced up by the thick lines circle) according to the rules.
(action example)
Below, the action example of the camera head 12b in the 3rd embodiment is described.The action of camera head 12b is except the contents processing of S27, and is identical with the action of camera head 12a in second embodiment.Therefore, about the action example of camera head 12b, only the contents processing to S27 describes.
Figure 18 is the part that the processing of the focusing object determination section 16b in the 3rd embodiment is shown, the i.e. process flow diagram of the contents processing of the S27 among Figure 13.In the 3rd embodiment, in S27, carry out the processing that this processing shown in Figure 180 is substituted in Figure 14 illustrated in second embodiment.
When beginning the processing of S27, division 22 is being categorized as a plurality of set (S31) by the detected a plurality of face points of face test section 17a.Then, set determination section 23 selects one to select set (S32) from a plurality of set that generated by division 22.
Center determination section 19b obtains the center (S33) of a plurality of face points that comprised in this selection set according to the face that is comprised in the selected selection set in S32.Then, face determination section 20b selects a face as the face (S34) that becomes the focusing object according to center obtained in S33.
(effect/effect)
According to the camera head 12b in the 3rd embodiment, in image, exist under the situation of a plurality of collectives, can be only according to the people who is comprised in one of them collective, decision becomes the face of focusing object.Under the user made a video recording the situation of certain collective, most cases was that the people of this collective concentrates on a place.At this moment, be not present in the frame even do not belong to people's (separately people, scattered distribution, other people's collective of people separately) of this collective, also expectation is subordinated to the subject that philtrum that the user wants the collective that makes a video recording is selected to the focusing object.According to the camera head 12b in the 3rd embodiment, can realize user's this expectation.That is, in the 3rd embodiment, be selected as making a video recording collective's (selecting set) of object of the collective's (set) that collects in the maximum people of quantity in the frame, realization is with the focusing of any artificial benchmark of belonging to this collective.In addition, as described in the variation hurdle of back, can constitute under the situation of the set that is not user expectation by the 23 selected selection set of set determination section, can manually select arbitrary collection as selecting set by the user.
Figure 19 is the figure that is used to represent by the obtained effect of camera head 12b.As shown in figure 19, even exist under the situation of a plurality of collectives,, select apart from the near face in the center of this collective (face that fences up by the thick lines rectangle) also only according to the position that for example is positioned at the face that collective comprised of image lower left.
(variation)
Set determination section 23 also can constitute according to the center of the face point of the face that is comprised in each set and decide the selection set.In this case, determination section 19b in center obtains the center at the face point that is comprised in each set that is generated by division 22.Set determination section 23 is according to the center by each obtained set of center determination section 19b, and set is selected in decision.For example, set determination section 23 is the set decision nearest apart from the central point of image of the centre coordinate of the face point that is comprised in the set to select set.And face determination section 20b is according to the center of selecting set, and decision becomes the face of focusing object.
And it is the face point that becomes the face of the face of focusing object that set determination section 23 also can constitute basis tentative decision of quilt in each set, determines to select set.In this case, determination section 18b is at each set that is generated by division 22, and only according to the face that is comprised in this set, tentative decision becomes the face of focusing object.And set determination section 23 is according to the face point in being gathered by obtained each of determination section 18b, and set is selected in decision.For example, set determination section 23 is the selection set to the set decision nearest apart from the central point of image of the coordinate of the face point of the tentative decision of quilt that is comprised in gathering.That is, gather the face that determination section 23 final decisions become the focusing object in fact.In this case, set determination section 23 also can constitute the face point coordinate of the face that is determined is passed to range finding portion 21.
And, under situation about constituting like this, also can constitute under the situation of set that by the 23 selected set of set determination section is not user expectation (selection result of set is not prompted to the user by making illustrated display part), use input part 13 to select other set as selecting set by the user.In this case, can constitute when input part 13 is operated, the set of being selected by the user just is shifted according to priority.This priority is based on the order of the stipulated standard of being judged by set determination section 23.
As mentioned above, face (being also referred to as " object face " sometimes) at the tentative decision focusing of each set object, the final object face of decision from tentative object face, this object face is prompted to the user, with the inconsistent situation of user's intention under, can change to the object face by in other object face of tentative decision any one.
Replace above-mentioned formation, can constitute at determination section 18b and from tentative object face, before the final object face of decision, tentative object face is prompted to the user, by the user from wherein selecting final object face.Tentative object face is not by making illustrated display part, and display box is prompted to the user around this face.
Perhaps, can before the decision objects face sets classification result of division 22 be prompted to the user by making illustrated display part, determine to select set by the user, determination section 18b is the alternative face from gathered by the selection that the user determined.In this case, the decision that set determination section 23 is not selected to gather is handled, and only carries out the determination result of user's selection set is passed to the action of determination section 18b.Like this, can come the decision action of opening/closing set determination section 23 by option operations.
[the 4th embodiment]
Below, the camera head 12c with focusing object determination section 16c is described.Below, at camera head 12c and focusing object determination section 16c, illustrate with the 3rd embodiment in camera head 12b and the difference of the object determination section 16b that focuses.
(system architecture)
Figure 20 is the figure of the functional block of expression camera head 12c.Camera head 12c has focusing object determination section 16c and replaces focusing object determination section 16b, and this point is different with camera head 12b in the 3rd embodiment.Therefore, below, explanation omitted to input part 13, image pickup part 14, image storage part 15 and the range finding portion 21 of camera head 12c.
<focusing object determination section 〉
Focusing object determination section 16c comprises that determination section 18c replaces determination section 18b, and this point is different with focusing object determination section 16b in the 3rd embodiment.Therefore, omission is to the explanation of face test section 17a, division 22 and set determination section 23.
<determination section 〉
Determination section 18c handles according to the face that is comprised in the set that is determined by set determination section 23 (selecting set).Determination section 18c is according to the size of this face, and decision becomes the face of focusing object.Determination section 18c is the face decision of the maximum in the face of for example selecting to be comprised in the set face that becomes the focusing object.Perhaps, determination section 18c also can be for example being determined to becoming the face of focusing object by the face with middle size in the face that is comprised in the set selection that determination section 23 the determined set.In the following description, establishing determination section 18c describes maximum face decision for the face that becomes the focusing object.
(action example)
Below, the action example of the camera head 12c in the 4th embodiment is described.The action of camera head 12c is except the contents processing of S27, and is identical with the action of camera head 12b in the 3rd embodiment.Therefore, about the action example of camera head 12c, only the contents processing to S27 describes.
Figure 21 is the part that the processing of the focusing object determination section 16c in the 4th embodiment is shown, the i.e. process flow diagram of the contents processing of the S27 among Figure 13.
When beginning the processing of S27, division 22 is being categorized as a plurality of set (S31) by the detected a plurality of face points of face test section 17a.Then, set determination section 23 selects one to select set (S32) from a plurality of set that generated by division 22.And determination section 18c selects a face as the face (S35) that becomes the focusing object according to the size of each face in the face that is comprised in selected selection set the processing of S32.
(effect/effect)
According to the camera head 12c in the 4th embodiment, the same with camera head 12b in the 3rd embodiment, in image, exist under the situation of a plurality of collectives, can be only according to the people who is comprised in one of them set, decision becomes the face of the object of focusing.
And,, the face that in a selected set, is comprised,, be selected to the face of focusing object according to the size of this face according to the camera head 12c in the 4th embodiment.Sometimes, want that for the user face of paying close attention to is apart from the nearest face of camera head most.And, be the possibility height of the face of maximum when making a video recording apart from the nearest face of camera head.Therefore, according to camera head 12c, the face that in a selected set, is comprised, select to be envisioned for according to the size of face the user is wanted the face paid close attention to most.Therefore, in this case, the user there is no need the face focusing to wanting to pay close attention to manually.At least, there is no need, wanting most that the face of paying close attention to is placed near the center of frame or its as conventional art.
(variation)
Determination section 18c also can constitute according to handle by the relevant characteristic quantity of each face that is comprised in the set that determined of set determination section 23.Determination section 18c obtains the characteristic quantity relevant with each face.Determination section 18c for example constitute size according to face, face towards etc., obtain the characteristic quantity relevant with each face.In addition, determination section 18c for example constitutes at each face, obtains representation class and is similar to the amount of degree of face as characteristic quantity." representation class is similar to the amount of the degree of face " for example uses the face at face test section 17a to detect when handling obtained value to represent.Specifically, when face test section 17a constitute judging whether comprise in certain zone calculate when having the face with this zone in the situation that is similar to the relevant value of face of image under, being similar to the relevant value of face as characteristic quantity with this.
[the 5th embodiment]
Below, the camera head 12d with focusing object determination section 16d is described.Below, at camera head 12d and focusing object determination section 16d, illustrate with second embodiment in camera head 12a and the difference of the object determination section 16a that focuses.
(system architecture)
Figure 22 is the figure of the functional block of expression camera head 12d.Camera head 12d has focusing object determination section 16d and replaces focusing object determination section 16a, and this point is different with camera head 12a in second embodiment.Therefore, below, explanation omitted to input part 13, image pickup part 14, image storage part 15 and the range finding portion 21 of camera head 12d.
<focusing object determination section 〉
Focusing object determination section 16d comprises face test section 17d, district's determination section 24 and determination section 18d.Below, each the function portion that constitutes focusing object determination section 16d is described.
<face test section 〉
Face test section 17d from focusing when detecting the face zone the image, detect face zone at each of a plurality of districts that are predetermined, this point is different with face test section 17a in second embodiment.Figure 23 is that expression is divided into the figure of the focusing in a plurality of districts with the example of image.As shown in figure 23, focusing with image be divided into a plurality of districts (in Figure 23, rectangular image (frame) is divided into 4 districts by the two-lines section of or transverse axis vertical with it and the central point by image, however be divided into what districts all can).For example will focus to be divided into which type of district with image by face test section 17d storage.
Face test section 17d detects face at each district in a plurality of districts.Then, face test section 17d passes to district's determination section 24 to testing result at each district.
<district determination section 〉
Being used in a plurality of districts that district's determination section 24 is selected to be predetermined determines to become the district (that is the district that, comprises the face that becomes the focusing object: hereinafter referred to as " selecting the district ") of the face of focusing object.That is, the coordinate that is delivered to range finding portion 21 is the face point coordinate of detected face in the selection district that is determined by district's determination section 24.
District's determination section 24 decides selection Qu Junke according to number, which benchmark such as characteristic quantity relevant with face detected in the district of face detected in each district.For example, district's determination section 24 is the district's decision that detects maximum faces in each district to select the district.Constituting like this under the situation of district's determination section 24, under situation shown in Figure 23, the district of upper left quarter is distinguished determination section 24 decisions for selecting the district.
And for example, district's determination section 24 also can be distinguished the face that is comprised in distinguishing district's decision relevant characteristic quantity and that be maximum with each for selection.As the computing method of the characteristic quantity relevant, can use the method identical with illustrated in the above-described embodiment method with face.
<determination section 〉
Determination section 18d is not the face that is comprised according in the integral image of being made a video recording by image pickup part 14, but handles according to the face that is comprised in the selection district that is determined by district's determination section 24, and this point is different with determination section 18a.Determination section 18d carries out the processing identical with determination section 18a except above-mentioned point.
Figure 24 is the figure of the processing example of expression determination section 18d.In district's (selecting the district), obtain the center of face point by district's determination section 24 selected upper left quarters.Then, according to obtained center, benchmark (being the near benchmark in central point apart from image (frame) here) is according to the rules selected a face point (the face point that is fenced up by the thick lines circle).
(action example)
Below, the action example of the camera head 12d in the 5th embodiment is described.The action of camera head 12d is except the contents processing of S19, and is identical with the action of camera head 12a in second embodiment.Therefore, about the action example of camera head 12d, only the contents processing to S19 describes.
Figure 25 is the part that the processing of the focusing object determination section 16d in the 5th embodiment is shown, the i.e. process flow diagram of the contents processing of the S19 among Figure 12.Like this, in the 5th embodiment, in S19, be substituted in applied processing shown in Figure 13 in second embodiment, and carry out processing shown in Figure 25.
When beginning the processing of S19, face test section 17d detects face (S36) at each district.Then, determination section 24 investigation in district's detect the number in the district of face.Be less than at the number in the district that detects face under 1 the situation that (S37:<1), promptly be under 0 the situation, district's determination section 24 is given determination section 18d the advisory that does not have face.Then, determination section 18d passes to range finding portion 21 (S38) to the two-dimensional coordinate of image (frame) central authorities.
(S37:>1), district's determination section 24 selects one to select to distinguish (S39) from a plurality of districts that detect face under more than 1 situation at the number in the district that detects face.Be (S37:=1) under one the situation in the district that detects face, perhaps after the processing of S39, determination section 18d investigation is by the number of the face that is comprised in the determination section 24 selected selection districts, district.(S40:>1), promptly detect under the situation of a plurality of faces in this selection district, determination section 18d selects a face (S41) from these a plurality of faces under more than 1 situation at the number of face.The processing of S41 is the processing identical with flow process shown in Figure 14.Yet process object is not integral image but selects the district that this point is different with second embodiment.
Get back to the explanation of using Figure 25.Be (S40:=1) under one the situation at detected face, perhaps after the processing of S41, determination section 18d passes to range finding portion 21 (S42) to the two-dimensional coordinate of the face point of a face.Here a said face is to be this unique face that is detected under one the situation at the number of detected face, is selecteed face under the situation after the processing of S41.After this is handled, implement S20 and processing afterwards among Figure 12.
(effect/effect)
According to the camera head 12d in the 5th embodiment, focusing is become a plurality of districts with image segmentation, from a plurality of districts, select to comprise a selection district of the face that becomes the focusing object.And, from this face of selecting to be selected to the district focusing object.Therefore, among the camera head 12d in the 5th embodiment, can obtain the effect identical similarly with the 3rd embodiment.Specifically, in image, exist under the situation of a plurality of collectives, can use rough these collectives that divide in district.And,, be selected to the face of the object of focusing according to the selected face of selecting the part of the collective that comprised in the district or collective of benchmark according to the rules.
And, different among the camera head 12d in the 5th embodiment with camera head 12b in the 3rd embodiment, not carrying out classification and handling.Therefore, among the camera head 12d in the 5th embodiment, compare with camera head 12b in the 3rd embodiment and can carry out fast processing.
(variation)
Determination section 18d also can be identical with the determination section 18b in the 3rd embodiment.In this case, determination section 18d constitutes and is not the face that comprised according in the selection set that is determined by set determination section 23, but handles according to the face that is comprised in the selection district that is determined by district's determination section 24, and this puts different with determination section 18b.That is, constitute be not with integral image but with select the district be object set classification and select the decision etc. of set.In this case, determination section 18d constitutes except above-mentioned point, carries out the processing identical with determination section 18b.
By such formation, there is no need to carry out with comprised in the picture to be had the face be the classification of object.That is, as long as only the face that is comprised in the selected district is carried out classification.Therefore, in picture, include under the situation of more a plurality of collectives, compare with camera head 12b in the 3rd embodiment and can carry out fast processing.And, compare with camera head 12d in common the 5th embodiment, can select to comprise the set of the face that becomes the focusing object more accurately and become the face of focusing object.
[the 6th embodiment]
Below, the camera head 12e with focusing object determination section 16e is described.Below, at camera head 12e and focusing object determination section 16e, illustrate with second embodiment in camera head 12a and the difference of the object determination section 16a that focuses.
(system architecture)
Figure 26 is the figure of the functional areas of expression camera head 12e.Camera head 12e has focusing object determination section 16e and replaces focusing object determination section 16a, and this point is different with camera head 12a in second embodiment.Therefore, below, explanation omitted to input part 13, image pickup part 14, image storage part 15 and the range finding portion 21 of camera head 12e.
<focusing object determination section 〉
Focusing object determination section 16e comprises face test section 17a, maximum face selection portion 25, candidate's selection portion 26 and determination section 18e.Below, each the function portion that constitutes focusing object determination section 16e is described.Wherein, the included face test section 17a structure of the included face test section 17a of focusing object determination section 16e and the focusing object determination section 16a in second embodiment is identical, thereby omits its explanation.
<maximum face selection portion 〉
Maximum face selection portion 25 is from by selecting maximum face (maximum face) the detected a plurality of faces of face test section 17a.Use under the situation of face rectangle in face test section 17a, maximum face selection portion 25 is selected according to the size of this face rectangle.And, in face test section 17a, to carry out under the situation of figure coupling, maximum face selection portion 25 is selected according to the size of this figure.Maximum face selection portion 25 constitutes other which kind of method of use is selected maximum face all can.Maximum face selection portion 25 passes to candidate's selection portion 26 to the information (for example size of the coordinate of face point, face) about selected maximum face.
Figure 27 is the figure that the processing example of maximum face selection portion 25 is shown.In Figure 27, the size of each face point is represented the size of the face corresponding with each face point.By the processing of maximum face selection portion 25, the face that is fenced up by thick frame clicked be selected as maximum face.
<candidate selection portion 〉
Candidate's selection portion 26 is selected to the candidate of the face of focusing object according to the size by maximum face selection portion 25 selected maximum faces.In other words, 26 decisions of candidate's selection portion become the face of the process object of determination section 18e.
Candidate's selection portion 26 is by maximum face selection portion 25 selected maximum faces with have the candidate who is chosen as the object face based on other face of the size in the specialized range of the size of this maximum face.Size in the specialized range is represented the size than the bashful for example several number percents of maximum~tens number percents.In addition, the size in the specialized range also represents for example to be the size more than the size or 2/3 over half of maximum face.
Figure 28 is the figure that the processing example of candidate's selection portion 26 is shown.In Figure 28, little open circles is represented not by the face point of candidate's selection portion 26 selections.The circle expression of blacking is by selected other the face point of face of candidate's selection portion 26.In addition, the face point that is surrounded by thick frame is represented the face point by the selected maximum face of determination section 18e.
<determination section 〉
Determination section 18e is not the face that is comprised according in the integral image of being made a video recording by image pickup part 14, but according to being handled by the face that is comprised in candidate's selection portion 26 selected candidates, this point is different with determination section 18a.Determination section 18e carries out the processing identical with determination section 18a except above-mentioned point.
Use Figure 28 that the processing of determination section 18e is described.Determination section 18e is selected to the face of focusing object according to by candidate's selection portion 26 selected three face points.At this moment, the center determination section 19e of determination section 18e obtains the center (circle of oblique line) of these three face points.And the face determination section 20e of determination section 18e is according to this center, and benchmark (being the nearest benchmark in center (central point) apart from image (frame) here) is selected a face point (being the face point (maximum face) that is fenced up by thick frame) here according to the rules.
(action example)
Below, the action example of the camera head 12e in the 6th embodiment is described.The action of camera head 12e is except the contents processing of S27, and is identical with the action of camera head 12a in second embodiment.Therefore, about the action example of camera head 12e, only the contents processing to S27 describes.
Figure 29 is the part that the processing of the focusing object determination section 16e in the 6th embodiment is shown, the i.e. process flow diagram of the contents processing of the S27 among Figure 13.Like this, the 6th embodiment is carried out processing shown in Figure 29 and is replaced processing shown in Figure 14 in S27, and this point is different with second embodiment.
When the processing of beginning S27, maximum face selection portion 25 is selected by the maximum face (S43) in the detected a plurality of faces of face test section 17a.Then, candidate's selection portion 26 is selected to candidate's face (S44) according to the size by maximum face selection portion 25 selected faces (maximum face).
Then, the center determination section 19e of determination section 18e obtains the center (S45) of face point according to becoming by candidate's selection portion 26 selected candidates' face (maximum face and at least one other face with the size in the specialized range that to belong to this maximum face be benchmark).Then, the face determination section 20e of determination section 18e selects a face according to obtained center, promptly becomes the face (S46) of focusing object.In addition, according to above-mentioned processing, under the situation that does not detect the face of working as, maximum face decision is the object face with other appearance.
(effect/effect)
According to the camera head 12e in the 6th embodiment,, be selected to the face of focusing object only according to face with size to a certain degree.Size to a certain degree is to judge according to the size of maximum face.
Therefore, can prevent the people that is taken as a setting, promptly the user has a mind to people's face as subject and is included in and is used as process object.Therefore, improved the set of selecting to comprise the face that becomes the focusing object processing, obtain processing, be selected to the precision of processing of the face of the object of focusing based on the center of the face that is comprised in the set.
(variation)
Maximum face selection portion 25 also can constitute according to the size in the zone of the colour of skin of detected face and select.
And second~the 6th embodiment also can constitute the face test section 5 that has in first embodiment and replace face test section 17a~17e.
The object determination device can be assembled on the video camera and camera that carries out based on the photography of focal length that is suitable for subject most or exposure etc.

Claims (44)

1. an object determination device is characterized in that, comprising:
Face detecting unit, the image that any focus information of its basis is made a video recording be as the process object image, from described process object image, detects people's face according to the relative value because of the statistic of the concavo-convex of people's face or a plurality of characteristic areas that organ produces; And
The decision unit, it determines subject of focusing and/or the subject that becomes the object of the control that exposes according to by the detected face of described face detecting unit when the enforcement shooting.
2. an object determination device is characterized in that, comprising:
Frame is obtained the unit, and it is from according to the sequential chart picture that comprises a plurality of frames that arbitrarily focus information is made a video recording, and the image of frame of obtaining regulation is as the process object image;
The face detecting unit, it detects people's face according to the relative value because of the statistic of the concavo-convex of people's face or a plurality of characteristic areas that organ produces from described process object image; And
The decision unit, it determines subject of focusing and/or the subject that becomes the object of the control that exposes according to by the detected face of described face detecting unit when the enforcement shooting.
3. object determination device according to claim 1 and 2, wherein, described face detecting unit is obtained in described process object image the relative value as the statistic of the first area of the characteristic area in the region-of-interest and second area, relative value according to described statistic judges whether include people's face in described region-of-interest, and then detects people's face.
4. according to any described object determination device in the claim 1~3, wherein, described decision unit detects under the situation of a plurality of faces at described face detecting unit, determine a face according to described a plurality of faces image information separately, according to this face described subject of focusing of decision and/or become the subject of the object of the control that exposes.
5. object determination device according to claim 4, wherein, described decision unit compares described a plurality of faces image size separately, decides described subject of focusing for maximum face and/or becomes the subject of the object of the control that exposes according to its size.
6. object determination device according to claim 4, wherein, described decision unit is at described each face, according to any or multinomial the counting the score in the density of the size of, face, position in image or face on every side of face, the described score of calculating is used as described image information.
7. according to any described object determination device in the claim 1~3, wherein, described decision unit detects under the situation of a plurality of people's faces at described face detecting unit, according to the result of gained that the material object of described a plurality of faces is found range respectively, according to its subject that decides described subject of focusing apart from face and/or become the object of the control that exposes for minimum.
8. a camera head is characterized in that, comprising:
Image unit, its shooting obtains the obtained focus information in unit and/or the image of the control information that exposes based on the focus information of regulation or by information;
The face detecting unit, it detects people's face according to the relative value because of the statistic of the concavo-convex of people's face or a plurality of characteristic areas that organ produces from the image of being made a video recording by described image unit according to the focus information of described regulation;
The decision unit, it determines subject of focusing and/or the subject that becomes the object of the control that exposes according to by the detected face of described face detecting unit when the enforcement shooting;
Information obtains the unit, and it is obtained and is used for the focus information that the subject that is determined by described decision unit is focused and/or is used for an exposure control information that the subject that is determined by described decision unit exposes and controls as object; And
Image storage unit, it is stored by described image unit according to obtained the image that obtained focus information in unit and/or exposure control information are made a video recording by described information.
9. a camera head is characterized in that, comprising:
Image unit, its according to the rules focus information or obtain obtained focus information in unit and/or exposure control information by information, shooting comprises the sequential chart picture of a plurality of frames;
Frame is obtained the unit, and it obtains the image of the frame of regulation from the sequential chart picture that comprises a plurality of frames of being made a video recording by described image unit;
The face detecting unit, it is from being obtained by described frame the obtained image in unit, detects people's face according to the relative value because of the statistic of the concavo-convex of people's face or a plurality of characteristic areas that organ produces;
The decision unit, it determines subject of focusing and/or the subject that becomes the object of the control that exposes according to by the detected face of described face detecting unit when the enforcement shooting;
Information obtains the unit, and it is obtained and is used for the focus information that the subject that is determined by described decision unit is focused and/or is used for an exposure control information that the subject that is determined by described decision unit exposes and controls as object; And
Image storage unit, its storage sequential chart picture, this sequential chart similarly are according to being obtained the obtained focus information in unit by described information and/or the exposure control information is made a video recording by described image unit.
10. according to Claim 8 or 9 described camera heads, also comprise range cells, it measures the distance from camera head to any subject,
Described decision unit detects under the situation of a plurality of people's faces at described face detecting unit,, decide described subject of focusing for minimum face and/or become the subject of the object of the control that exposes find range the respectively result of gained of the material object of described a plurality of faces according to described range cells according to its distance.
11. according to Claim 8 or 9 described camera heads, also comprise display unit, people's face that it is determined by described decision unit and the difference of other faces demonstration of coming.
12. one kind is used to make signal conditioning package to carry out the program of following steps:
The image of making a video recording according to any focus information as the process object image, from described process object image, is detected the step of people's face according to the relative value because of the statistic of the concavo-convex of people's face or a plurality of characteristic areas that organ produces; And
According to detected face, the step of the subject of subject that decision is focused when the enforcement shooting and/or the object that becomes the control that exposes.
13. an object determination device is characterized in that, comprising:
Detecting unit, it detects people's face from the image of being imported; And
The decision unit, when being detected a plurality of people's face by described detecting unit, it is according to the position of this a plurality of people's faces, and decision becomes the face of the object of focusing and/or becomes the face of the object controlled of exposing when implementing shooting from these a plurality of people's faces.
14. an object determination device is characterized in that, comprising:
Frame is obtained the unit, and its image of frame of obtaining regulation from the sequential chart picture of being imported that comprises a plurality of frames is as the process object image;
Detecting unit, it detects people's face from described process object image; And
The decision unit, when being detected a plurality of people's face by described detecting unit, it is according to the position of this a plurality of people's faces, and decision becomes the face of the object of focusing and/or becomes the face of the object controlled of exposing when implementing shooting from these a plurality of people's faces.
15. according to claim 13 or 14 described object determination devices, described decision unit comprises:
Decision unit, center, it determines the center of the position of described a plurality of people's faces according to the position of described a plurality of people's faces; And
Face decision unit, it is according to the face of the described object of described center decision becoming.
16. object determination device according to claim 15, wherein, described center determines the external polygonal center decision in the unit handle and the position of described a plurality of people's faces to be described center.
17. object determination device according to claim 15, wherein, decision unit, described center is described center to the decision of the center of gravity of the position of described a plurality of people's faces.
18. object determination device according to claim 15, wherein, described face decision unit is the people face decision nearest apart from described center the face that becomes described object.
19. object determination device according to claim 15, wherein, described face decision unit is benchmark according to the rules, from being arranged in apart from the face of the described object of face decision becoming of described center gauge set a distance.
20. according to claim 13 or 14 described object determination devices, wherein, described decision unit the face decision that is positioned at below in described a plurality of people's faces for becoming the face of described object.
21. an object determination device is characterized in that, comprising:
Detecting unit, it detects people's face from the image of being imported; And
The decision unit, when detecting a plurality of people's face by described detecting unit, its number with detected face is a benchmark, is the people's face decision that is positioned at its center the face that becomes the face of the object of focusing and/or become the object of the control that exposes when implementing shooting.
22. an object determination device is characterized in that, comprising:
Frame is obtained the unit, and its image of frame of obtaining regulation from the sequential chart picture of being imported that comprises a plurality of frames is as the process object image;
Detecting unit, it detects people's face from described process object image; And
The decision unit, when detecting a plurality of people's face by described detecting unit, its number with detected face is a benchmark, is the people's face decision that is positioned at its center the face that becomes the face of the object of focusing and/or become the object of the control that exposes when implementing shooting.
23. any described object determination device according in the claim 13,14,21,22 also comprises:
Taxon, when being detected a plurality of people's face by described detecting unit, it is categorized as a plurality of set with detected a plurality of people's faces according to separately position; And
Set decision unit, they are from described a plurality of set, and set is selected in decision, and this choice set share the face that becomes described object in decision,
Described decision unit is according to the people's face that is comprised in the described selection set, and decision becomes the face of described object.
24. any described object determination device according in the claim 13,14,21,22 is characterized in that, also comprises display unit, it distinguishes the demonstration of coming by people's face and other faces that described decision unit is determined.
25. an object determination device is characterized in that, comprising:
Detecting unit, it detects people's face from the image of being imported;
Taxon, when being detected a plurality of people's face by described detecting unit, its detected a plurality of people's faces are categorized as a plurality of set;
Tentative decision unit, it determines the face that becomes the face of the object of focusing and/or become the object of the control that exposes when implementing shooting at each set in described a plurality of set people's face that is comprised in this set; And
The final decision unit, its final decision from the face that is determined by described tentative decision unit becomes the face of described object.
26. an object determination device is characterized in that, comprising:
Frame is obtained the unit, and its image of frame of obtaining regulation from the sequential chart picture of being imported that comprises a plurality of frames is as the process object image;
Detecting unit, it detects people's face from described process object image;
Taxon, when being detected a plurality of people's face by described detecting unit, its detected a plurality of people's faces are categorized as a plurality of set;
Tentative decision unit, it determines the face that becomes the face of the object of focusing and/or become the object of the control that exposes when implementing shooting at each set in described a plurality of set people's face that is comprised in this set; And
The final decision unit, its final decision from the face that is determined by described tentative decision unit becomes the face of described object.
27. an object determination device is characterized in that, comprising:
Detecting unit, it has carried out cutting apart and has detected people's face in each district in a plurality of districts of obtaining at the image that will be imported;
District decision unit, it is according to the testing result of described detecting unit, and the district is selected in decision, and this selections district is used to determine to become the face of the object of focusing when implementing to make a video recording and/or becomes the face of the object controlled of exposing; And
The decision unit, decision becomes the face of described object its face that is comprised in described selection district.
28. an object determination device is characterized in that, comprising:
Frame is obtained the unit, and its image of frame of obtaining regulation from the sequential chart picture of being imported that comprises a plurality of frames is as the process object image;
Detecting unit detects people's face in each district in a plurality of districts that it has obtained in that described process object image has been carried out cutting apart;
District decision unit, it is according to the testing result of described detecting unit, and the district is selected in decision, and this selections district is used to determine to become the face of the object of focusing when implementing to make a video recording and/or becomes the face of the object controlled of exposing; And
The decision unit, decision becomes the face of described object its face that is comprised in described selection district.
29. an object determination device is characterized in that, comprising:
Detecting unit, it detects people's face from the image of being imported;
Judging unit, when detecting a plurality of people's face by described detecting unit, the face of the maximum in people's face that its judgement is detected;
Selected cell, it is selected the face of described maximum and has the face of the size in the specialized range that size with this maximum face is a benchmark from described detected face; And
The decision unit, it determine to become the face of the object of focusing when implementing shooting and/or becomes the face of the object controlled of exposing from described selecteed face.
30. an object determination device is characterized in that, comprising:
Frame is obtained the unit, and its image of frame of obtaining regulation from the sequential chart picture of being imported that comprises a plurality of frames is as the process object image;
Detecting unit, it detects people's face from described process object image;
Judging unit, when being detected a plurality of people's face by described detecting unit, it judges the face of the maximum in detected people's face;
Selected cell, it is selected the face of described maximum and has the face of the size in the specialized range that size with this maximum face is a benchmark from described detected face; And
The decision unit, it determine to become the face of the object of focusing when implementing shooting and/or becomes the face of the object controlled of exposing from described selected face.
31. an object determination device is characterized in that, comprising:
Detecting unit, it detects people's face from the image of being imported;
Taxon, when being detected a plurality of people's face by described detecting unit, it is categorized as a plurality of set with detected a plurality of people's faces according to separately position;
Set decision unit, it determines to select set from described a plurality of set, and this choice set share the face that becomes the face of the object of focusing and/or become the object of the control that exposes when implementing shooting in determining; And
The decision unit, it is from being become the face of described object by decision the face that is comprised in the described set decision selection that the unit the determined set.
32. an object determination device is characterized in that, comprising:
Frame is obtained the unit, and its image of frame of obtaining regulation from the sequential chart picture of being imported that comprises a plurality of frames is as the process object image;
Detecting unit, it detects people's face from described process object image;
Taxon, when being detected a plurality of people's face by described detecting unit, it is categorized as a plurality of set with detected a plurality of people's faces according to separately position;
Set decision unit, it determines to select set from described a plurality of set, and this choice set share the face that becomes the face of the object of focusing and/or become the object of the control that exposes when implementing shooting in determining; And
The decision unit, it is from being become the face of described object by decision the face that is comprised in the described set decision selection that the unit the determined set.
33. one kind is used to make signal conditioning package to carry out the program of following steps:
From the image of being imported, detect the step of people's face; And
When detecting a plurality of people's face, according to the position of these a plurality of people's faces, decision becomes the face of the object of focusing and/or becomes the step of face of the object of the control that exposes when implementing shooting from these a plurality of people's faces.
34. one kind is used to make signal conditioning package to carry out the program of following steps:
From the image of being imported, detect the step of people's face; And
When detecting a plurality of people's face, be benchmark with the number of detected face, the people's face decision that is positioned at its center for when implementing shooting, becoming the face of the object of focusing and/or becoming the step of face of the object of the control that exposes.
35. one kind is used to make signal conditioning package to carry out the program of following steps:
From the image of being imported, detect the step of people's face;
When detecting a plurality of people's face, detected a plurality of people's faces are categorized as the step of a plurality of set according to position separately;
At each set in described a plurality of set, decision becomes the face of the object of focusing and/or becomes the step of the face of the object controlled of exposing the people's face that is comprised in this set when implementing shooting; And
Final decision becomes the step of the face of described object in the face that is determined from described each set.
36. one kind is used to make signal conditioning package to carry out the program of following steps:
Carried out cutting apart and detected the step of people's face in each district in a plurality of districts of obtaining at the image that will be imported;
According to the testing result in described each district, the step in district is selected in decision, and this selections district is used to determine to become the face of the object of focusing when implementing to make a video recording and/or becomes the face of the object controlled of exposing; And
Decision becomes the step of the face of described object the face that is comprised in described selection district.
37. one kind is used to make signal conditioning package to carry out the program of following steps:
From the image of being imported, detect the step of people's face;
When detecting a plurality of people's face, judge the step of the face of the maximum in a plurality of people's faces that are detected;
From the described face that is detected, select the face of described maximum and have the step of the face of the size in the specialized range that size with this maximum face is a benchmark; And
Decision becomes the face of the object of focusing and/or becomes the step of face of the object of the control that exposes when implementing shooting from described selecteed face.
38. one kind is used to make signal conditioning package to carry out the program of following steps:
From the image of being imported, detect the step of people's face;
When detecting a plurality of people's face, detected a plurality of people's faces are categorized as the step of a plurality of set according to position separately;
The step of set is selected in decision from described a plurality of set, and this choice set share the face that becomes the face of the object of focusing and/or become the object of the control that exposes when implementing shooting in determining; And
Decision becomes the step of the face of described object the face that is comprised in the described selection set that is determined.
39. object determination device according to claim 1 and 2, it is characterized in that, when detecting a plurality of people's face by described face detecting unit, described decision unit determines the face that becomes the face of the object of focusing and/or become the object of the control that exposes when implementing shooting according to the position of these a plurality of people's faces from these a plurality of people's faces.
40. object determination device according to claim 1 and 2, it is characterized in that, when detecting a plurality of people's face by described face detecting unit, described decision unit is a benchmark with the number of detected face, is the people's face decision that is positioned at its center the face that becomes the face of the object of focusing and/or become the object of the control that exposes when implementing shooting.
41. object determination device according to claim 1 and 2 is characterized in that, described decision unit comprises:
Taxon, when being detected a plurality of people's face by described face detecting unit, its detected a plurality of people's faces are categorized as a plurality of set;
Tentative decision unit, it determines the face that becomes the face of the object of focusing and/or become the object of the control that exposes when implementing shooting at each set in described a plurality of set people's face that is comprised in this set; And
The final decision unit, its final decision from the face that is determined by described tentative decision unit becomes the face of described object.
42. object determination device according to claim 1 and 2, it is characterized in that, also comprise decision unit, district, the district is selected in its testing result decision according to described face detecting unit, this selection district is used to determine the face that becomes the face of the object of focusing and/or become the object of the control that exposes when implementing shooting
Decision becomes the face of described object the face that described decision unit is comprised in described selection district.
43. object determination device according to claim 1 and 2 is characterized in that, also comprises:
Judging unit, when detecting a plurality of people's face by described face detecting unit, the face of the maximum in people's face that its judgement is detected; And
Selected cell, it is selected the face of described maximum and has the face of the size in the specialized range that size with this maximum face is a benchmark from the described face that is detected,
Described decision unit determines the face that becomes the face of the object of focusing and/or become the object of the control that exposes when implementing shooting from described selecteed face.
44. object determination device according to claim 1 and 2 is characterized in that, also comprises:
Taxon, when being detected a plurality of people's face by described face detecting unit, it is categorized as a plurality of set with detected a plurality of people's faces according to separately position; And
Set decision unit, it determines to select set from described a plurality of set, and this choice set share the face that becomes the face of the object of focusing and/or become the object of the control that exposes when implementing shooting in determining,
Described decision unit is from being become the face of described object by decision the face that is comprised in the described set decision selection that the unit the determined set.
CNB200480020302XA 2003-07-15 2004-07-14 Object determination device and camera head Expired - Fee Related CN100538498C (en)

Applications Claiming Priority (3)

Application Number Priority Date Filing Date Title
JP2003197404 2003-07-15
JP197404/2003 2003-07-15
JP318701/2003 2003-09-10

Publications (2)

Publication Number Publication Date
CN1823298A true CN1823298A (en) 2006-08-23
CN100538498C CN100538498C (en) 2009-09-09

Family

ID=36923837

Family Applications (1)

Application Number Title Priority Date Filing Date
CNB200480020302XA Expired - Fee Related CN100538498C (en) 2003-07-15 2004-07-14 Object determination device and camera head

Country Status (1)

Country Link
CN (1) CN100538498C (en)

Cited By (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101848337A (en) * 2009-03-25 2010-09-29 索尼公司 Image processing equipment, image processing method and program
CN101141608B (en) * 2007-09-28 2011-05-11 腾讯科技(深圳)有限公司 Video instant communication system and method
CN101242494B (en) * 2007-01-17 2012-03-21 三星电子株式会社 Digital photographing apparatus and method for controlling the same
US8384792B2 (en) 2007-11-02 2013-02-26 Sony Corporation Imaging apparatus, method for controlling the same, and program
US8730375B2 (en) 2007-05-18 2014-05-20 Casio Computer Co., Ltd. Imaging apparatus having focus control function
CN105227831A (en) * 2015-09-06 2016-01-06 Tcl集团股份有限公司 A kind of method and system of self adaptation zoom
CN107948522A (en) * 2017-12-04 2018-04-20 广州视源电子科技股份有限公司 Method, device, terminal and storage medium for selecting shot person head portrait by camera
CN108076278A (en) * 2016-11-10 2018-05-25 阿里巴巴集团控股有限公司 A kind of Atomatic focusing method, device and electronic equipment
CN108289161A (en) * 2017-01-09 2018-07-17 三星电子株式会社 Electronic equipment and its image capture method
CN111198549A (en) * 2020-02-18 2020-05-26 陈文翔 Poultry breeding monitoring management system based on big data
CN111263072A (en) * 2020-02-26 2020-06-09 Oppo广东移动通信有限公司 Shooting control method and device and computer readable storage medium
CN111758117A (en) * 2018-03-14 2020-10-09 欧姆龙株式会社 Inspection system, recognition system, and learning data generation device

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7298412B2 (en) * 2001-09-18 2007-11-20 Ricoh Company, Limited Image pickup device, automatic focusing method, automatic exposure method, electronic flash control method and computer program
JP2003189296A (en) * 2001-12-21 2003-07-04 Konica Corp Photographing apparatus

Cited By (20)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101242494B (en) * 2007-01-17 2012-03-21 三星电子株式会社 Digital photographing apparatus and method for controlling the same
US8730375B2 (en) 2007-05-18 2014-05-20 Casio Computer Co., Ltd. Imaging apparatus having focus control function
CN103905729A (en) * 2007-05-18 2014-07-02 卡西欧计算机株式会社 Imaging device and program thereof
CN103905729B (en) * 2007-05-18 2018-01-16 卡西欧计算机株式会社 Camera device and focusing controlling method and recording medium
CN101141608B (en) * 2007-09-28 2011-05-11 腾讯科技(深圳)有限公司 Video instant communication system and method
US8384792B2 (en) 2007-11-02 2013-02-26 Sony Corporation Imaging apparatus, method for controlling the same, and program
CN101848337A (en) * 2009-03-25 2010-09-29 索尼公司 Image processing equipment, image processing method and program
CN101848337B (en) * 2009-03-25 2012-07-04 索尼公司 Image processing device, image processing method, and program
CN105227831A (en) * 2015-09-06 2016-01-06 Tcl集团股份有限公司 A kind of method and system of self adaptation zoom
CN105227831B (en) * 2015-09-06 2019-08-06 Tcl集团股份有限公司 A kind of method and system of adaptive zoom
CN108076278A (en) * 2016-11-10 2018-05-25 阿里巴巴集团控股有限公司 A kind of Atomatic focusing method, device and electronic equipment
CN108289161A (en) * 2017-01-09 2018-07-17 三星电子株式会社 Electronic equipment and its image capture method
US10871798B2 (en) 2017-01-09 2020-12-22 Samsung Electronics Co., Ltd. Electronic device and image capture method thereof
CN107948522A (en) * 2017-12-04 2018-04-20 广州视源电子科技股份有限公司 Method, device, terminal and storage medium for selecting shot person head portrait by camera
CN107948522B (en) * 2017-12-04 2020-06-26 广州视源电子科技股份有限公司 Method, device, terminal and storage medium for selecting shot person head portrait by camera
CN111758117A (en) * 2018-03-14 2020-10-09 欧姆龙株式会社 Inspection system, recognition system, and learning data generation device
CN111758117B (en) * 2018-03-14 2023-10-27 欧姆龙株式会社 Inspection system, recognition system, and learning data generation device
CN111198549A (en) * 2020-02-18 2020-05-26 陈文翔 Poultry breeding monitoring management system based on big data
CN111198549B (en) * 2020-02-18 2020-11-06 湖南伟业动物营养集团股份有限公司 Poultry breeding monitoring management system based on big data
CN111263072A (en) * 2020-02-26 2020-06-09 Oppo广东移动通信有限公司 Shooting control method and device and computer readable storage medium

Also Published As

Publication number Publication date
CN100538498C (en) 2009-09-09

Similar Documents

Publication Publication Date Title
CN1675919A (en) Imaging system and image processing program
CN1213592C (en) Adaptive two-valued image processing method and equipment
CN1219395C (en) Camera apparatus and camera method
CN1203662C (en) Automatic grey-degree grade compensation apparatus, method and recording medium thereof
CN1261811C (en) Photograph picture printer and printing method, program and printing medium unit
CN100346352C (en) Image display apparatus and short film generation apparatus
CN1288174A (en) Apparatus and method for taking face photography
CN1867940A (en) Imaging apparatus and image processing method therefor
CN1324873C (en) Boundary detection method between areas having different features in image data
CN1823298A (en) Object decision device and imaging device
CN1809116A (en) Image processing apparatus, image forming apparatus, image reading process apparatus, image processing method
CN101052989A (en) Image processing unit, mask generation method, and program
CN1882036A (en) Image processing apparatus and method
CN1599406A (en) Image processing method and device and its program
CN1805531A (en) Information processing apparatus, information processing method, recording medium, and program
CN1764228A (en) Image processing apparatus, image forming apparatus, method for processing image,
CN1691740A (en) Magnified display apparatus and magnified image control apparatus
CN1607551A (en) Method and apparatus for image-based photorealistic 3D face modeling
CN1846232A (en) Object posture estimation/correlation system using weight information
CN1839410A (en) Image processor, imaging apparatus and image processing method
CN1940965A (en) Information processing apparatus and control method therefor
CN1645241A (en) Imaging apparatus, method and device for processing images
CN1684491A (en) Image processing device and method, image projection apparatus, and program
CN1580934A (en) Picture frame width regulating device, shooting device and developing and printing device
CN1645941A (en) Image processing apparatus, image pickup apparatus, image processing method, image data output method

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C14 Grant of patent or utility model
GR01 Patent grant
CF01 Termination of patent right due to non-payment of annual fee
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20090909