JP2016114785A - Imaging device, imaging method and imaging program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To accurately focus on a subject.SOLUTION: An imaging device 10 comprises an imaging unit 20 that includes an image pick-up element 19 having a phase difference detection unit detecting a phase difference for calculating a focused position by an image plane phase difference AF arranged, and photographs a subject via a lens 12 capable of moving a focus position.The imaging device 10 also classifies a plurality of estimation windows 24 set so as to include mutually different phase difference detection units into a plurality of groups on the basis of a magnitude relation between the phase differences detected by the phase difference detection unit to be included in the individual estimation window 24, and comprises a specification unit 40 that specifies a main subject area 28 in accordance with a magnitude relation between a variation of a position of the estimation window 24 to be included in the classified individual group and a preliminarily set threshold and a calculation unit 50 that calculates a focused position on the basis of an average of the phase differences included in the main subject area 28 specified by the specification unit 40 and detected by the phase difference detection unit.SELECTED DRAWING: Figure 3

Description

  The disclosed technology relates to an imaging apparatus, an imaging method, and an imaging program.

  In recent years, a camera incorporating an auto focus (AF) technique for detecting a lens position (focus position) for capturing an image and focusing on a subject and moving the lens to the detected focus position, and Smartphones are popular.

  In order to detect the in-focus position, it is necessary to measure the distance from the camera or the like to the subject. Therefore, a plurality of photoelectric conversion cells arranged on the light receiving surface of the image sensor are used to acquire reflected light from the subject and detect the amount of light that changes according to the position of the photoelectric conversion cell, thereby measuring the distance to the subject. Distance technology is disclosed.

JP 2003-214847 A JP 2009-3261 A

  However, the conventional distance measurement technology does not take into account noise included in the output of the image sensor, which is caused by, for example, variations in characteristics of the image sensor and temperature changes around the image sensor. An error may occur in the distance to the subject.

  As one aspect, the disclosed technique aims to accurately focus on a subject even when noise is superimposed on an output of an image sensor.

  In one aspect, an imaging apparatus includes an imaging device having a phase difference detection unit that detects a phase difference for calculating a focus position by an image plane phase difference detection method, and an optical system that can move a focal position An image pickup unit that picks up an image of a subject with light incident on the light receiving surface of the image pickup device. In addition, the imaging apparatus may select a plurality of first regions set to include different phase difference detection units based on the magnitude relationship of the phase differences detected by the phase difference detection units included in the individual first regions. Classify into multiple groups. Then, the imaging apparatus specifies the second region that is estimated to correspond to the subject according to the magnitude relationship between the variation in the position of the first region included in each classified group and a preset threshold value. A part. In addition, the imaging apparatus includes a calculation unit that calculates an in-focus position based on an average of the phase differences detected by the phase difference detection unit included in the second region specified by the specifying unit.

  As one aspect, the disclosed technique has an effect that the subject can be focused with high accuracy even when noise is superimposed on the output of the image sensor.

It is a schematic diagram explaining a focus. It is a schematic diagram explaining the principle of focusing in an image plane phase difference AF method. It is an example of the functional block diagram which shows the function of an imaging device. It is a figure which shows the example of arrangement | positioning of a photoelectric conversion cell. It is a figure which shows an example of a ranging area. It is a figure which shows an example of an estimation window. It is a schematic diagram explaining the classification | category method of an estimation window. It is a schematic diagram explaining the average and the dispersion | variation in the position of an estimation window. It is a schematic diagram explaining a main subject area. FIG. 25 is a diagram illustrating an example of a configuration when an imaging apparatus is realized by a computer. It is a flowchart which shows an example of the flow of an imaging process. It is a schematic diagram explaining the calculation method of the phase difference of an estimation window. It is a flowchart which shows an example of the flow of main subject area | region selection processing. It is a figure which shows an example of a rotation angle table. It is a figure which shows an example of a captured image. It is a schematic diagram explaining the classification | category method of another estimation window.

  Hereinafter, an example of an embodiment of the disclosed technology will be described in detail with reference to the drawings. In addition, the same code | symbol is attached | subjected to the component and process which a function bears the same function through all the drawings, and the overlapping description may be abbreviate | omitted suitably.

  First, an AF method used in an imaging apparatus having a function of capturing an image such as a smartphone or a digital camera will be described.

  In an imaging apparatus, for example, a contrast AF method and an image plane phase difference AF method are often used as an AF method.

  The contrast AF method focuses on the subject by using the principle that the difference between the contrast of the image, that is, the brightness in the darkest part of the image and the brightness in the brightest part of the image is the largest at the in-focus position. It is a method to match. Specifically, the contrast of the subject is calculated while moving the lens, and the lens position where the contrast of the subject is highest, that is, the focus position is detected.

  On the other hand, the image plane phase difference AF method uses the positional deviation (also referred to as phase difference) of the image of the subject obtained by each of the reflected light at a specific portion of a plurality of locations of the subject among the reflected light from the subject, This method focuses on the subject. Hereinafter, the reflected light may be simply referred to as “light”.

  Here, the phase difference of the image of the subject in the imaging apparatus will be described.

  FIG. 1 is a diagram illustrating a state of a subject image captured by the imaging apparatus. Light from the subject is refracted by the lens 12, and an image of the subject is formed on the light receiving surface of the image sensor. An image formed on the light receiving surface of the image sensor by light from the subject is referred to as a formed image, and hereinafter simply referred to as “image”.

  On the light receiving surface of the image sensor, photoelectric conversion cells are arranged two-dimensionally in the left-right direction and the vertical direction of the light receiving surface, and the image sensor receives the amount of light received by each photoelectric conversion cell. An image is picked up by outputting an electric signal corresponding to the pixel value of the image.

  As shown in FIG. 1, when the lens 12 is in the in-focus position, for example, the light R that passes through the right part of the lens 12 and the light L that passes through the left part of the lens 12 are received on the light receiving surface. The position matches. Therefore, the image by the light R and the image by the light L overlap, and an image with high sharpness is obtained. In this case, the phase difference between the subject image by the light R and the subject image by the light L is “0”.

  On the other hand, in the light incident direction, when the lens 12 is closer to the subject than the in-focus position, the image of the light R on the light receiving surface and the light L are compared to when the lens 12 is in the in-focus position. Are formed at positions apart from each other. Therefore, compared with the case where the lens 12 is in the in-focus position, the pixel value output from the photoelectric conversion cell changes gently, so the image is blurred and the sharpness of the image is lowered. In this way, a state where the lens 12 is closer to the subject than the in-focus position is referred to as defocus (front). In some cases, the subject image by the light R is called a right image, and the subject image by the light L is called a left image.

  Further, when the lens 12 is located farther from the subject than the in-focus position in the light incident direction, the image is compared with the case where the lens 12 is in the in-focus position for the same reason as the defocus (previous). The sharpness is low. In this way, the state in which the lens 12 is located farther from the subject than the in-focus position is referred to as defocus (rear).

  That is, the subject can be focused by moving the lens 12 to a position where the distance between the right image and the left image on the light receiving surface, that is, the phase difference between the right image and the left image is as close to “0” as possible.

  Therefore, in the image plane phase difference AF method, in order to form a right image and a left image on the light receiving surface, for example, a microlens is disposed on the front surface of the light receiving surface of the image sensor. The light R that has passed through the right part of the lens 12 passes through the right side of the microlens, and the light L that has passed through the left part of the lens 12 passes through the left side of the microlens. There are a plurality of types of microlenses having different light transmission ranges.

  Among the microlenses, there is a microlens that blocks light on the left side of the microlens and transmits light incident on the right side of the microlens, that is, light R. In addition, among the microlenses, for example, there is a microlens that transmits light that is incident on the left side portion of the microlens, that is, the light L, by blocking the right side portion of the microlens. Furthermore, among the microlenses, there is a microlens that allows light incident on any part of the microlens to pass through without being blocked.

  FIG. 2 is a diagram for schematically explaining the principle of focusing in the imaging apparatus using the image plane phase difference AF method. The left portion 14RL of the micro lens 14R is shielded from light, and the right portion 14RR transmits the light R. Further, the right side portion 14LR of the micro lens 14L is shielded from light, and the left side portion 14LL transmits the light L.

  In this case, the photoelectric conversion cell (phase difference detection cell) that has received the light R transmitted through the microlens 14R outputs the pixel value of each pixel included in the right image. The photoelectric conversion cell (phase difference detection cell) that receives the light L that passes through the microlens 14L outputs the pixel value of each pixel included in the left image.

  The pixel value of the right image in FIG. 2 indicates an output when a plurality of photoelectric conversion cells in a predetermined column aligned in the left-right direction of the light receiving surface receive the light R for convenience of explanation. The pixel value of the left image indicates an output when a plurality of photoelectric conversion cells included in the same column as the photoelectric conversion cell that outputs the pixel value of the right image receive the light L.

  Then, the imaging apparatus calculates the phase difference between the right image and the left image, and moves the lens 12 by a correction distance for correcting the phase difference to “0” according to the calculated phase difference, thereby aligning the lens 12. Place it near the focal position to focus on the subject.

  Thus, in the image plane phase difference AF method, the lens 12 can be moved from the current position to the in-focus position using the correction distance obtained from the phase difference between the right image and the left image. In comparison, the subject may be focused at high speed. On the other hand, in the contrast AF method, the contrast of the subject is calculated at the position of each lens 12 while continuously moving the lens 12, so that the time required to focus on the subject compared to the image plane phase difference AF method. May become longer. However, as already described, the contrast AF method detects the position of the lens 12 at which the contrast is highest while continuously moving the lens 12, so that the subject is more accurate than the image plane phase difference AF method. It may be possible to focus.

  Therefore, the imaging apparatus according to the present embodiment employs an AF method that combines the contrast AF method and the image plane phase difference AF method and takes advantage of the mutual method. That is, the image pickup apparatus according to the present embodiment moves the lens 12 near the in-focus position by the image plane phase difference AF method, and then continuously moves the lens 12 in the vicinity of the in-focus position by the contrast AF method. Further, the position of the lens 12 in focus on the subject is detected.

  However, noise may be superimposed on the pixel value output from the photoelectric conversion cell of the image sensor due to variations in characteristics due to individual differences between the photoelectric conversion cells of the image sensor and the influence of the temperature around the image sensor. In the image plane phase difference AF method, since the correction distance of the lens 12 is calculated using the phase difference between the right image and the left image, as the noise superimposed on the pixel value output from the photoelectric conversion cell of the image sensor increases, The focusing accuracy of the image pickup apparatus tends to decrease.

  Therefore, hereinafter, an imaging device that can focus on a subject with high accuracy even when noise is superimposed on a pixel value output from a photoelectric conversion cell will be described.

  FIG. 3 is an example of a functional block diagram illustrating functions of the imaging apparatus 10 according to the present embodiment.

  The imaging device 10 includes an imaging unit 20, a setting unit 30, a specifying unit 40, a calculating unit 50, a driving unit 60, and a contrast AF unit 70.

  The imaging unit 20 includes an imaging device, receives light that passes through the lens 12 and enters the light receiving surface of the imaging device, and images an electrical signal output from the imaging device according to the amount of received light Is output to the setting unit 30 as a pixel value.

  FIG. 4 is a diagram illustrating an example of a photoelectric conversion cell arranged on the light receiving surface of the image sensor.

  As shown in FIG. 4, a plurality of photoelectric conversion cells are arranged two-dimensionally on the light receiving surface of the image sensor. Note that the photoelectric conversion cell includes the types of the photoelectric conversion cell 16, the photoelectric conversion cell 17, and the photoelectric conversion cell 18.

  The photoelectric conversion cell 16 is a photoelectric conversion cell that receives light from a microlens that transmits light without blocking light. Therefore, the photoelectric conversion cell 16 is used for imaging a subject because light from the subject is received without being blocked.

  The photoelectric conversion cell 17 is a photoelectric conversion cell (phase difference detection cell) that blocks light on the right side of the light receiving surface and receives light from the microlens 14L that transmits light L incident on the left side of the light receiving surface. Therefore, the photoelectric conversion cell 17 is used for capturing a left image of the subject. Hereinafter, in order to clarify the type of photoelectric conversion cell, the photoelectric conversion cell 17 is particularly referred to as “phase difference detection cell 17”.

  The photoelectric conversion cell 18 is a photoelectric conversion cell (phase difference detection cell) that receives light from the microlens 14R that shields the left portion of the light receiving surface and transmits the light R incident on the right portion of the light receiving surface. Therefore, the photoelectric conversion cell 18 is used for capturing the right image of the subject. Hereinafter, in order to clarify the type of the photoelectric conversion cell, the photoelectric conversion cell 18 is particularly referred to as a “phase difference detection cell 18”.

  As shown in FIG. 4, the imaging element has a light receiving surface in which the photoelectric conversion cell 16 is partially replaced with a phase difference detection cell 17 and a phase difference detection cell 18. The phase difference detection cell 17 and the phase difference detection cell 18 are arranged in pairs on the light receiving surface. In FIG. 4, shaded portions of the phase difference detection cell 17 and the phase difference detection cell 18 indicate light shielding portions.

  The setting unit 30 in FIG. 3 includes a ranging area setting unit 32 and a phase difference calculation area setting unit 34, and the pixel value of the image output from the imaging unit 20 is input to the ranging area setting unit 32.

  The ranging area setting unit 32 sets a ranging area on the image. The distance measurement area is an area used for focusing, and the imaging apparatus 10 focuses on the subject using the pixel values of the pixels included in the distance measurement area. That is, the imaging device 10 focuses on the subject included in the distance measurement area.

  FIG. 5 is a diagram illustrating an example of a distance measurement area. Here, for convenience of explanation, the image is represented by replacing the photoelectric conversion cell 16, the phase difference detection cell 17, and the phase difference detection cell 18 that correspond one-to-one with the image pixel. The same expression is used in FIG. 6 described later.

  In FIG. 5, the distance measurement area 22 is set at the center of the image, but the setting position of the distance measurement area 22 is not limited.

  As shown in FIG. 6, the phase difference calculation area setting unit 34 in FIG. 3 sets a plurality of estimation windows 24 within the distance measurement area 22 set by the distance measurement area setting unit 32. The estimation window 24 is an area provided for calculating the phase difference from the left image detected by the phase difference detection cell 17 included therein and the right image detected by the phase difference detection cell 18. is there. Therefore, the phase difference calculation region setting unit 34 sets the estimation window 24 so as to include a plurality of pairs of the phase difference detection cell 17 and the phase difference detection cell 18.

  Using the estimation window 24 set in the distance measurement area 22 by the phase difference calculation area setting unit 34, the specifying unit 40 is estimated to include the main subject that is the object of imaging from the distance measurement area 22. Identify the area. Therefore, the specification unit 40 includes a phase difference acquisition unit 42, a classification unit 44, a variation calculation unit 46, and a main subject region specification unit 48.

  The phase difference acquisition unit 42 calculates the phase difference between the right image and the left image in each estimation window 24, calculates the average value (average phase difference) of the phase differences in each estimation window 24, and outputs it to the classification unit 44. Output.

  The classification unit 44 classifies the plurality of estimation windows 24 set in the distance measurement area 22 into an estimation window 24 whose phase difference is less than the average phase difference and an estimation window 24 whose phase difference is equal to or greater than the average phase difference. The classification result is output to the variation calculation unit 46. For convenience of explanation, a group of estimation windows 24 whose phase difference is less than the average phase difference is referred to as S group, and a group of estimation windows 24 whose phase difference is equal to or greater than the average phase difference is referred to as L group.

  FIG. 7 is a diagram schematically illustrating the function of the classification unit 44.

A graph 26 illustrated in FIG. 7 is a diagram illustrating a distribution example of the phase difference when O is the origin, the phase difference is taken on the horizontal axis, and the number of estimation windows 24 is taken on the vertical axis. W ₀ represents an average phase difference. In this case, the estimation window 24 having a phase difference less than the average phase difference W ₀ is classified into the S group, and the estimation window 24 having a phase difference greater than or equal to the average phase difference W ₀ is classified into the L group. Of the estimation windows 24, the estimation window 24 classified into the S group is referred to as an estimation window 24S, and the estimation window 24 classified into the L group is referred to as an estimation window 24L. Further, hereinafter, the estimation window 24 is used as a generic meaning of the estimation window 24S and the estimation window 24L.

The variation calculation unit 46 in FIG. 3 calculates the average of the positions of the estimation windows included in each group from the arrangement of the estimation windows 24S and the estimation windows 24L in the ranging area 22 and the S and L groups. The position variation is calculated. Hereinafter, the average of the positions of the estimation windows 24S in the S group is represented as μ _S , and the variation in the positions of the estimation windows 24S in the S group is represented as σ _S. In addition, the average position of the estimation window 24L in the L group is denoted by μ _L , and the variation in the position of the estimation window 24L in the L group is denoted by σ _L.

  FIG. 8 is a diagram schematically illustrating the function of the variation calculation unit 46. In FIG. 8, for convenience of explanation, the S group estimation window 24 </ b> S and the L group estimation window 24 </ b> L included in the ranging area 22 are separately illustrated.

  The positions of the estimation window 24S and the estimation window 24L are represented by XY coordinates with one vertex P of the distance measurement area 22 as the origin, for example. It should be noted that there is no restriction on which part of the estimation window 24S and the estimation window 24L is the position of the estimation window 24S and the estimation window 24L on the XY coordinates. The definition of the position of the window 24S and the estimation window 24L may be the same. For example, the barycentric points of the estimation window 24S and the estimation window 24L, or the upper right point (for example, the point P1) of the estimation window 24S and the estimation window 24L may be determined as the positions of the estimation windows 24S and the estimation windows 24L.

  In addition, although point P1 exists in each estimation window 24 shown in FIG. 8, since drawing becomes complicated, point P1 is illustrated with respect to some estimation windows 24. FIG.

3 calculates the average μ _S of the position of the estimation window 24S, the variation σ _S of the position of the estimation window 24S, and the average μ _L of the position of the estimation window 24L from the position of each estimation window 24. And the variation σ _L of the position of the estimation window 24L is calculated and output to the main subject region specifying unit 48.

The main subject region specifying unit 48 estimates the higher probability that the image of the main subject is included from the S group and the L group according to the magnitude relationship between the predetermined threshold T and the variations σ _S and σ _L. A group of windows 24 is selected. A specific method for selecting the group of the estimation window 24 will be described later.

  Further, the main subject region specifying unit 48 sets a closed region having a predetermined shape including each of the estimated windows 24 belonging to the selected classification in the image, and positions of the estimated windows 24 of the selected group included in the closed region. And the variation σ of the position of the estimation window 24 are calculated.

  Then, the main subject region specifying unit 48 specifies a region in the range of μ ± σ centering on the average μ in the distance measuring region 22 as a region including the main subject. Note that an area considered to include the main subject in the distance measurement area 22 is referred to as a main subject area.

  FIG. 9 is a diagram showing the main subject area set by the main subject area specifying unit 48. As an example, the main subject area 28 when the main subject area specifying unit 48 selects the S group is shown. FIG. 9 shows a case where the closed region 27 including each of the estimation windows 24S included in the S group and the main subject region 28 are the same region, but depending on the values of the average μ and the variation σ. The closed region 27 and the main subject region 28 may indicate different ranges.

  Then, the main subject area specifying unit 48 in FIG. 3 outputs the specified main subject area 28 to the calculating unit 50.

The calculation unit 50 calculates the average value μ ₀ of the phase difference in each of the estimation windows 24 included in the main subject region 28. Then, the calculation unit 50 calculates the rotation angle of the motor corresponding to the correction distance that is the movement distance of the lens 12 necessary to bring the average value μ ₀ close to 0, and outputs the calculated rotation angle to the drive unit 60. .

  The drive unit 60 drives the motor according to the rotation angle calculated by the calculation unit 50, moves the lens 12 from the current position to the in-focus position, and focuses on the subject. Then, the driving unit 60 notifies the contrast AF unit 70 that the focusing process by the image plane phase difference AF method has been completed.

  The contrast AF unit 70 further adjusts the in-focus position of the lens 12 using a known focus AF method.

  Next, FIG. 10 shows a configuration diagram when the imaging apparatus 10 is realized by a computer.

  The computer 100 includes a CPU 102, a memory 104, and a nonvolatile storage unit 106. The CPU 102, the memory 104, and the nonvolatile storage unit 106 are connected to each other via a bus 108. The computer 100 also includes an I / O (Input / Output) 110 for connecting other devices and the computer 100 and transmitting / receiving data to / from each other, and the I / O 110 is connected to the bus 108.

  For example, an input device 112, a display device 114, an image sensor 19, and a motor 116 are connected to the I / O 110. The input device 112 includes devices for giving instructions to the imaging device 10 by the user, such as various switches, buttons, and a touch panel. The input device 112 includes a reading device for reading data recorded on the recording medium 150 such as a CD-ROM or a flash memory. The display device 114 includes, for example, a display device such as a display that displays the state of the imaging device 10 and the captured image to the user. The image sensor 19 converts the amount of light received through the lens 12 into an electrical signal in accordance with an instruction from the computer 100. The motor 116 changes the position of the lens 12 connected via a gear or the like in accordance with an instruction from the computer 100. The storage unit 106 can be realized by an HDD (Hard Disk Drive), a flash memory, or the like.

  The storage unit 106 stores an imaging program 120 for causing the computer 100 to function as the imaging device 10 illustrated in FIG. The imaging program 120 stored in the storage unit 106 includes a ranging area setting process 122, a phase difference calculation area setting process 124, a phase difference acquisition process 126, a classification process 128, a variation calculation process 130, and a main subject area specifying process 132. Including. Further, the imaging program 120 includes a calculation process 134, a driving process 136, and a contrast AF process 138.

  The CPU 102 reads out the imaging program 120 from the storage unit 106 and develops it in the memory 104, and executes each process included in the imaging program 120.

  The CPU 102 reads out the imaging program 120 from the storage unit 106, expands it in the memory 104, and executes the imaging program 120, whereby the computer 100 operates as the imaging device 10 illustrated in FIG. 3. Further, when the CPU 102 executes the ranging area setting process 122, the computer 100 operates as the ranging area setting unit 32 shown in FIG. Further, when the CPU 102 executes the phase difference calculation area setting process 124, the computer 100 operates as the phase difference calculation area setting unit 34 illustrated in FIG. Further, when the CPU 102 executes the phase difference acquisition process 126, the computer 100 operates as the phase difference acquisition unit 42 illustrated in FIG. Further, when the CPU 102 executes the classification process 128, the computer 100 operates as the classification unit 44 illustrated in FIG. Further, when the CPU 102 executes the variation calculation process 130, the computer 100 operates as the variation calculation unit 46 illustrated in FIG. Further, when the CPU 102 executes the main subject region specifying process 132, the computer 100 operates as the main subject region specifying unit 48 shown in FIG. Further, when the CPU 102 executes the calculation process 134, the computer 100 operates as the calculation unit 50 illustrated in FIG. Further, when the CPU 102 executes the driving process 136, the computer 100 operates as the driving unit 60 shown in FIG. Furthermore, when the CPU 102 executes the contrast AF process 138, the computer 100 operates as the contrast AF unit 70 shown in FIG.

  The CPU 102 executes the distance measurement area setting process 122 and the phase difference calculation area setting process 124 so that the computer 100 operates as the setting unit 30 illustrated in FIG. Further, when the CPU 102 executes the phase difference acquisition process 126, the classification process 128, the variation calculation process 130, and the main subject region specifying process 132, the computer 100 operates as the specifying unit 40 illustrated in FIG.

  Further, the CPU 102 develops the threshold value included in the threshold value storage area 140 in the memory 104, whereby the threshold value T for classifying the variation in the position of the estimation window 24 is stored in the memory 104. Further, the CPU 102 expands the image sensor information included in the image sensor information storage area 142 in the memory 104, whereby the image sensor arrangement table is stored in the memory 104. The imaging element arrangement table is a table showing the arrangement of the photoelectric conversion cells 16, the phase difference detection cells 17, and the phase difference detection cells 18 on the light receiving surface of the imaging element 19, and each pixel of the image is any of the pixels. Whether the pixel corresponds to the type of cell can be acquired.

  The computer 100 can be realized by, for example, a semiconductor integrated circuit, more specifically, an ASIC (Application Specific Integrated Circuit) or the like.

  Next, the operation of the imaging device 10 according to the present embodiment will be described. The imaging device 10 according to the present embodiment performs imaging processing when an instruction to start imaging is received from the input device 112, for example, in response to a user operation.

  FIG. 11 is a flowchart illustrating an example of the flow of imaging processing of the imaging apparatus 10 according to the present embodiment.

  First, in step S <b> 10, the ranging area setting unit 32 acquires pixel values output from each cell of the image sensor 19. Then, the distance measurement area setting unit 32 stores each acquired pixel value in a predetermined area of the memory 104 in association with the position on the light receiving surface of the cell that outputs the pixel value. Thus, a set of pixel values stored in the memory 104 becomes an image.

  Next, the ranging area setting unit 32 sets the ranging area 22 on the image. In the example of FIG. 5, the ranging area setting unit 32 sets the ranging area 22 in a rectangular shape, but the shape of the ranging area 22 is not limited. Further, as already described, there is no restriction on the set position of the distance measurement area 22 on the image. However, the distance measurement area setting unit 32 may set the distance measurement area 22 in an area where it is estimated that some object, such as an animal, a plant, or a building, is captured using, for example, a known image recognition technique. preferable.

  In step S20, the phase difference calculation area setting unit 34 sets a plurality of estimation windows 24 in the distance measurement area 22 set in the process of step S10 as shown in FIG. When the phase difference calculation area setting unit 34 sets the estimation window 24, each estimation window 24 includes a pair of a pixel corresponding to the phase difference detection cell 17 and a pixel corresponding to the phase difference detection cell 18. Each estimation window 24 is set so as to be This is because the phase difference acquisition unit 42 uses the left image obtained from the pixel value of the pixel corresponding to the phase difference detection cell 17 and the pixel value of the pixel corresponding to the phase difference detection cell 18 in each of the estimation windows 24. This is because the right image obtained can be acquired.

  Note that the phase difference calculation region setting unit 34 may select pixels corresponding to the phase difference detection cell 17 and the phase difference detection cell 18 with reference to the image sensor arrangement table stored in advance in the memory 104.

  In step S30, the phase difference acquisition unit 42 determines whether or not the phase difference between the right image and the left image has been calculated in each of the estimation windows 24 set in the process of step S20. If the determination is negative, the process proceeds to step S40.

  In step S40, the phase difference acquisition unit 42 selects one unselected estimation window 24 from each of the estimation windows 24 set in the process of step S20. Then, the phase difference acquisition unit 42 reads the pixel value of the pixel corresponding to the phase difference detection cell 17 included in the selected estimation window 24 and acquires the left image, and also includes the phase difference included in the selected estimation window 24. The pixel value of the pixel corresponding to the detection cell 18 is read to acquire the right image.

  Next, in step S50, the phase difference acquisition unit 42 moves the right image acquired in the process of step S40 pixel by pixel within a predetermined number of pixels along the left-right direction of the image. Then, the phase difference acquisition unit 42 calculates a sum of absolute differences (SAD: Sum of Absolute Difference) between the right image and the left image at each position for each movement. Specifically, the sum of absolute differences between the right image and the left image is the pixel of the pixel corresponding to the phase difference detection cell 17 for each of the phase difference detection cell 17 and the phase difference detection cell 18 that form a pair. It is represented by the sum of the absolute values of the differences between the value and the pixel value of the pixel corresponding to the phase difference detection cell 18.

  Then, the phase difference acquisition unit 42 refers to the sum of absolute differences at each position within a predetermined number of pixels, and moves from the position before the movement of the right image to the position where the sum of absolute differences is smallest. The number of pixels is calculated as a phase difference between the right image and the left image.

  Then, the phase difference acquisition unit 42 associates the calculated phase difference with the information of the estimation window 24 that is the phase difference calculation source, stores the phase difference in a predetermined area of the memory 104, and estimates the phase difference calculation source. The selected identifier is set in the information of the window 24. Accordingly, in the processing of step S30 and step S40, whether or not the estimation window 24 has calculated the phase difference is determined according to the presence or absence of the selected identifier set in the estimation window 24.

  The predetermined number of pixels represents a maximum value of the number of moving pixels when moving the right image, and is a numerical value obtained in advance by an experiment with an actual apparatus of the imaging apparatus 10 or a computer simulation based on a design specification of the imaging apparatus 10. is there.

  Further, in this embodiment, the example in which the position of the left image is fixed and the right image is moved has been described. However, even if the position of the right image is fixed and the left image is moved, the same difference absolute value sum is obtained. It goes without saying that it is obtained.

  FIG. 12 is a diagram schematically showing the process of step S50. As shown in FIG. 12, by moving the right image in the direction of the arrow AR indicating the left-right direction of the image within a predetermined number of pixels, a graph 52 showing the sum of absolute differences for each moving position is obtained. It is done. The origin Q represents the position of the right image before the movement. For example, the phase difference when the right image is moved in the AR1 direction is represented by “+”, and the phase difference when the right image is moved in the AR2 direction is represented by “−”.

  On the other hand, if the determination in step S30 is affirmative, that is, if the processing in steps S40 and S50 is executed for each estimation window 24 and the phase difference in each estimation window 24 is calculated, step S60 is performed. Migrate to

  In step S <b> 60, the specifying unit 40 performs a main subject area selection process for selecting the main subject area 28 from the distance measurement area 22. Accordingly, the main subject area selection process will be described below, and after the description is finished, the description returns to the imaging process in FIG.

  FIG. 13 is a flowchart showing an example of the flow of the main subject area selection process executed in the process of step S60.

  First, in step S600, the phase difference acquisition unit 42 reads the phase difference of each estimation window 24 calculated in the process of step S50 from the memory 104, and calculates the average phase difference of all the estimation windows 24 that read the phase difference. To do. Then, the phase difference acquisition unit 42 stores the calculated average phase difference in a predetermined area of the memory 104.

  In step S610, the classification unit 44 reads from the memory 104 the phase difference of each estimation window 24 calculated in the process of step S50 and the average phase difference calculated in the process of step S600. Then, the classification unit 44 sets the plurality of estimation windows 24 set in the ranging area 22 in the process of step S20, the estimation window 24 of the S group whose phase difference is less than the average phase difference, and the phase difference equal to or greater than the average phase difference. And the estimation window 24 of the L group. The classification unit 44 associates an identifier indicating the S group with the estimation window 24 </ b> S classified into the S group, and associates an identifier indicating the L group with the estimation window 24 </ b> L classified into the L group. Is stored in a predetermined area. An identifier indicating whether the estimation window 24 is the S group or the L group is referred to as a classification identifier.

  As described above, the reason why the estimation window 24 is classified into the S group and the L group is to cope with the case where the distance measurement area 22 includes a main subject and a subject (non-main subject) that is not the main subject. Here, as an example, the classification unit 44 classifies the estimation window 24 into a plurality of types using the average phase difference. The reason why the average phase difference is adopted as a reference used for classification of the estimation window 24 is that the main subject and the non-main subject tend to be separated with the average phase difference as a boundary.

  In step S620, the variation calculation unit 46 acquires the position of each estimation window 24 when a predetermined point of the distance measurement area 22 is set as the origin, as indicated by a vertex P shown in FIG. For example, the variation calculation unit 46 defines the upper right point P1 of each estimation window 24 as the position of each estimation window 24 in the XY coordinates with the vertex P of the ranging area 22 as the origin, and the position of the estimation window 24 To get.

Then, the variation calculation unit 46 refers to the classification identifier of the estimation window 24 set in the process of step S610, and the average μ _S of the positions of the estimation windows 24S included in the S group, and the variation σ _S of the positions of the estimation windows 24S. Is calculated. Further, the variation calculation unit 46 refers to the classification identifier of the estimation window 24 set in the process of step S610, and calculates the average μ _L of the position of the estimation window 24L and the variation σ _L of the position of the estimation window 24L.

  As an example, the variation calculation unit 46 calculates the variation in the position of the estimation window 24 as a standard deviation, but is not limited thereto. The variation calculation unit 46 may use any index as long as it can quantitatively represent the variation in the position of the estimation window 24. For example, the variation calculation unit 46 may use the distance from the average of the positions of the estimation windows 24 included in each group to the estimation window 24 farthest as the variation in the positions of the estimation windows 24 included in the group. Further, the variation calculation unit 46 may use the variance value as the variation in the position of the estimation window 24.

The variation calculation unit 46 stores the calculated average μ _S , variation σ _S , average μ _L , and variation σ _L in a predetermined area of the memory 104.

In step S <b> 630, the main subject region specifying unit 48 acquires the threshold value T, the average μ _S , the variation σ _S , the average μ _L , and the variation σ _L from the memory 104. Then, the main subject region specifying unit 48 determines whether or not the variation σ _S and the variation σ _L are both smaller than the threshold T. If the determination is affirmative, the process proceeds to step S640. If the determination is negative, step S650 is performed. Migrate to

The threshold value T is a value given by a function f ₁ (σ ₀ ) having a predetermined standard deviation σ ₀ as an argument, and affects the estimation of the imaging position of the main subject in the distance measurement area 22. Defines the reference amount of noise to be superimposed on the pixel value. The standard deviation σ ₀ and the function f ₁ (σ ₀ ) are obtained in advance by an experiment using the actual apparatus of the imaging apparatus 10 or a computer simulation based on the design specifications of the imaging apparatus 10, and are stored in advance in a predetermined area of the memory 104, for example. It is a value.

Next, the main subject area specifying unit 48 estimates the imaging position of the main subject in the distance measurement area 22 based on the magnitude relationship between the variation σ _S and the variation σ _L and the threshold value T.

For example, when the variation σ _S and the variation σ _L are both smaller than the threshold value T, it is considered that the noise superimposed on the pixel value of the image is less than the reference amount. Since the main subject tends to be photographed larger than the non-main subject, in such a situation where the noise is less than the reference value, the main subjects are included in the group having the larger variations σ _S and σ _L among the S group and the L group. There is a high possibility that it exists in the distribution range of the estimated window 24.

Therefore, in step S640, the main subject region specifying unit 48 selects a group having a larger variation σ _S , σ _L (one having a wider distribution range of the estimation window 24) out of the S group and the L group.

On the other hand, when at least one of the variation σ _S and the variation σ _L is equal to or greater than the threshold T, it is highly likely that the noise has an influence, and the noise superimposed on the pixel value of the image is in a state larger than the reference amount. It is done. In a situation where such noise is larger than the reference value, the distribution range of the estimation window 24 included in each group of the S group and the L group includes an error with respect to the distribution range of the subject on the corresponding image. Accordingly, it is difficult for the main subject region specifying unit 48 to correctly determine the position of the main subject from the size relationship between the main subject and the non-main subject. However, the main subject tends to exist in the vicinity of the average position of the estimation window 24. Therefore, in a situation where there is more noise than the reference value, there is a high possibility that the main subject is present in the distribution range of the estimation window 24 included in the group with the smaller variation of the S group and the L group.

  Therefore, in step S650, the main subject region specifying unit 48 selects a group having a smaller variation (a distribution range of the estimation window 24 is narrower) out of the S group and the L group.

  In step S660, the main subject area specifying unit 48 determines the closed area 27 having a predetermined shape and the minimum size that includes each of the estimation windows 24 belonging to the group selected in the process of step S640 or step S650. Set to. Then, in order to reflect the biased state of the position of the selected estimation window 24, the main subject area specifying unit 48 reflects the average μ of each position of the selected estimation window 24 and the position of the estimation window 24 in the closed area 27. The variation σ is calculated.

That is, as shown in FIG. 9, the closed region 27 may include both the estimation window 24S included in the S group and the estimation window 24L included in the L group. However, the main subject region specifying unit 48 uses only the estimation window 24 selected in Step S640 or Step S650 among the estimation windows 24 included in the closed region 27, and uses the average μ of the positions of the estimation windows 24 in the closed region 27. And a variation σ of the position of the estimation window 24 is calculated.
Note that the average μ _S of the position of the estimation window 24S and the variation σ _S of the position of the estimation window 24S, or the variation of the average μ _L of the position of the estimation window 24L and the position of the estimation window 24L, which have already been calculated in the process of step S620. σ _L may be used.

  Then, the main subject region specifying unit 48 specifies a region in the range of μ ± σ around the average μ as the main subject region 28 in the distance measuring region 22. When the main subject area 28 is set so as to include a part of the estimation window 24, the main subject area 28 is reset so as to include the entire estimation window 24 including the part or the part is included. The main subject area 28 is reset so as to remove the estimated window 24. This is because it is necessary to calculate the average of the phase difference in units of the estimation window 24 in the process of step S70 in FIG.

  As described above, the main subject area selection process in step S60 shown in FIG. 11 is completed, and the main subject area 28 representing the imaging position of the main subject is selected from the distance measurement area 22.

  The description returns to the imaging process of the imaging apparatus 10 shown in FIG.

In step S70, the calculation unit 50 calculates an average value μ ₀ of the phase difference in each of the estimation windows 24 included in the main subject region 28 selected in the process of step S60. As the phase difference in each of the estimation windows 24 included in the main subject region 28, the phase difference of each estimation window 24 that has already been calculated by the phase difference acquisition unit 42 by the process in step S50 can be used.

Since the average value μ ₀ is a value indicating the amount of defocus of the estimation window 24 at the imaging position of the main subject, the lens 12 is moved so that the average value μ ₀ approaches 0, so that the focus of the lens 12 is adjusted. Can be adjusted to the main subject.

Therefore, the calculation unit 50 refers to the rotation angle table that associates the average value μ ₀ of the phase difference with the rotation angle of the motor 116 for moving the lens 12 by the correction distance corresponding to the average value μ _0. The rotation angle of the motor 116 is acquired. Note that the rotation angle table is obtained in advance by an experiment with an actual apparatus of the imaging apparatus 10 or a computer simulation based on the design specification of the imaging apparatus 10 and is stored in a predetermined area of the memory 104.

  FIG. 14 shows an example of the rotation angle table 80. Each value of the illustrated rotation angle table 80 is a value for explaining the rotation angle table 80 and is different from a value actually used in the imaging device 10. The sign of the rotation angle in the rotation angle table 80 represents the rotation direction of the motor 116. For example, a positive rotation angle indicates that the motor 116 is rotated by an angle represented by the absolute value of the rotation angle in the rotation direction of the motor 116 that moves the lens 12 in a direction to approach the light receiving surface. The negative rotation angle indicates that the motor 116 is rotated by an angle represented by the absolute value of the rotation angle in the rotation direction of the motor 116 that moves the lens 12 away from the light receiving surface.

  Then, the calculation unit 50 stores the rotation angle acquired from the rotation angle table 80 in a predetermined area of the memory 104.

The average value μ ₀ is the average of the phase differences calculated from the estimation window 24 included in the main subject region 28 regardless of the group to which the estimation window 24 belongs, and thus is calculated from the estimation window 24 included in one group. Compared to the average value of the phase differences, the amount of focus shift with respect to the main subject is accurately represented.

In the processes of the step S70, the calculating unit 50 has been calculated rotation angle corresponding to the average value mu ₀ with the rotation angle table 80, the function f _{2 (μ 0)} of the average value mu ₀ and argument e.g. It goes without saying that the rotation angle corresponding to the average value μ ₀ may be calculated using.

  In step S80, the drive unit 60 acquires the rotation angle of the motor 116 calculated in the process of step S70 from the memory 104. And the drive part 60 drives the motor 116 according to the acquired rotation angle.

  As described above, the lens 12 moves to the vicinity of the in-focus position of the main subject by the imaging processing from step S10 to S80 in FIG. Thereafter, the contrast AF unit 70 adjusts the focusing position of the lens 12 using a focusing technique based on a known contrast AF method, so that the in-focus state is adjusted only by the image plane phase difference AF method. Improves focus accuracy.

  Here, as shown in FIG. 15, imaging in the case where a person 72 as a main subject and a vehicle 74 as a non-main subject located behind the person 72 are included in the distance measurement area 22 will be considered.

  15 includes a person 72 and a vehicle 74 having different subject distances. Therefore, if the correction distance is simply calculated from the average phase difference of each estimation window 24 included in the distance measurement area 22, there may be a situation in which no subject is focused.

  However, in the imaging apparatus 10 according to the present embodiment, the main subject area 28 that is more likely to include the image of the main subject than the other areas is identified from the distance measurement area 22, and the main subject area 28 is identified. A correction distance is calculated from the average phase difference of each estimation window 24 included. Therefore, the imaging apparatus 10 can focus on the main subject with higher accuracy than when the correction distance is calculated from the average phase difference of the estimation windows 24 included in the distance measurement area 22. For example, in the case of FIG. 15, the imaging apparatus 10 can focus on the person 72.

  Furthermore, in the imaging apparatus 10 according to the present embodiment, the estimation window 24 included in the ranging area 22 is divided into the S group and the L group, the degree of variation in the position of the estimation window 24 in each group, the threshold T, and the like. Depending on the size relationship, a group that is more likely to contain the image of the main subject than one group is selected. That is, in the imaging device 10 according to the present embodiment, it is determined which group is likely to contain the image of the main subject according to the amount of noise superposition included in the pixel value output by the photoelectric conversion cell. . Therefore, for example, the subject can be more accurately focused as compared to a case where it is fixedly determined that the image of the main subject is included in the region including the group with the larger degree of variation in the position of the estimation window 24. it can.

In the process of step S610 illustrated in FIG. 13, the classification unit 44 calculates the average of the estimation windows 24 included in the ranging area 22 and the estimation windows 24 in which the phase difference of the estimation window 24 is included in the ranging area 22. They were classified into the S group and L groups according to whether less than a phase difference W _0. However, the classification method of the estimation window 24 included in the distance measurement area 22 is not limited to this.

FIG. 16 is a diagram obtained by extracting the graph 26 from FIG. The graph 26 illustrated, two peak values W ₂ and W ₃ are present in phase difference. In such a case, it can be considered that one peak value represents the peak value of the phase difference distribution for the main subject, and the other peak value represents the peak value of the phase difference distribution for the non-main subject. Therefore, the classification unit 44 uses the phase difference W ₁ at the minimum point existing between the peak value W ₂ and the peak value W ₃ as a reference value, and uses the estimation window 24 included in the distance measurement area 22 as the estimation window 24. it may be classified into a group S and L groups phase difference depending on whether less than a phase difference W ₁ of.

  Although the disclosed technology has been described using the embodiments, the disclosed technology is not limited to the scope described in each embodiment. Various modifications or improvements can be added to each embodiment without departing from the spirit of the disclosed technique, and forms to which the modifications or improvements are added are also included in the technical scope of the disclosed technique. For example, the processing order may be changed without departing from the scope of the disclosed technology.

  In the embodiment, the aspect in which the imaging program 120 is stored (installed) in the storage unit in advance has been described, but the present invention is not limited to this. The imaging program 120 according to the disclosed technique can also be provided in a form recorded on a computer-readable recording medium 150. For example, the imaging program 120 according to the disclosed technology can be provided in a form recorded on a portable recording medium such as a CD-ROM, a DVD-ROM, and a USB memory. The imaging program 120 according to the disclosed technique can also be provided in a form recorded in a semiconductor memory such as a flash memory.

  Regarding the above embodiment, the following additional notes are disclosed.

(Appendix 1)
The phase difference detection unit for detecting the phase difference for calculating the in-focus position by the image plane phase difference detection method includes imaging elements arranged at a plurality of positions in the light receiving surface, and through an optical system that can move the focal position. An imaging unit that images a subject with light incident on a light receiving surface of the imaging element;
A plurality of first regions set so as to include the phase difference detection units different from each other are grouped into a plurality of groups based on the magnitude relationship of the phase differences detected by the phase difference detection units included in the individual first regions. And a specifying unit that specifies a second region estimated to correspond to the subject according to a magnitude relationship between a variation in position of the first region included in each classified group and a preset threshold value When,
A calculating unit that calculates the in-focus position based on an average of phase differences detected by the phase difference detecting unit included in the second region specified by the specifying unit;
An imaging apparatus including:

(Appendix 2)
In the case where each variation in the position of the first region included in the individual group is smaller than the preset threshold value, the specifying unit has a larger variation in the position of the first region in the individual group. When at least one of the variations in the position of the first region included in the individual group is larger than the preset threshold, the position of the first region in the individual group is selected. The imaging apparatus according to claim 1, wherein a group having a smaller variation is selected, and the second region is identified from an average value and a variation of each position of the first region belonging to the selected group.

(Appendix 3)
The specifying unit is within a range represented by a variation in the position of the first region belonging to the selected group, based on a point represented by an average value of the position of the first region belonging to the selected group. The imaging device according to claim 2, wherein the region is specified as the second region.

(Appendix 4)
The identifying unit compares each phase difference of the first region with an average value of each phase difference of the first region, and classifies each of the first regions into the plurality of groups. The imaging apparatus according to any one of 1 to appendix 3.

(Appendix 5)
The imaging device according to any one of Supplementary Note 1 to Supplementary Note 4, wherein the degree of variation in position of the first region is expressed by a standard deviation.

(Appendix 6)
On the computer,
Through an optical system that can move the focal position by image sensors with multiple phase difference detectors that detect the phase difference for calculating the in-focus position using the image plane phase difference detection method. Image a subject represented by light incident on a light receiving surface of the image sensor;
A plurality of first regions set so as to include the phase difference detection units different from each other are grouped into a plurality of groups based on the magnitude relationship of the phase differences detected by the phase difference detection units included in the individual first regions. And specifying a second region that is estimated to correspond to the subject according to a magnitude relationship between a variation in position of the first region included in each classified group and a preset threshold value,
An imaging method for executing a process including calculating the in-focus position based on an average of phase differences detected by the phase difference detection unit included in the identified second region.

(Appendix 7)
When each variation in position of the first region included in the individual group is smaller than the preset threshold value, a group having a larger variation in position of the first region is selected from the individual groups, If at least one of the variations in the position of the first region included in the individual group is greater than the preset threshold, the group in which the variation in the position of the first region is smaller among the individual groups The imaging method according to claim 6, wherein the second region is specified from an average value and variation of positions of the first regions belonging to the selected group.

(Appendix 8)
A region within a range represented by a variation in the position of the first region belonging to the selected group, based on a point represented by an average value of the positions of the first region belonging to the selected group, The imaging method according to appendix 7, which is specified as the second region.

(Appendix 9)
The phase difference of each of the first regions is compared with the average value of the phase difference of each of the first regions, and each of the first regions is classified into the plurality of groups. The imaging method according to any one of the above.

(Appendix 10)
The imaging method according to any one of appendix 6 to appendix 9, wherein the degree of variation in the position of the first region is expressed by a standard deviation.

(Appendix 11)
On the computer,
Through an optical system that can move the focal position by image sensors with multiple phase difference detectors that detect the phase difference for calculating the in-focus position using the image plane phase difference detection method. Image a subject represented by light incident on a light receiving surface of the image sensor;
A plurality of first regions set so as to include the phase difference detection units different from each other are grouped into a plurality of groups based on the magnitude relationship of the phase differences detected by the phase difference detection units included in the individual first regions. And specifying a second region that is estimated to correspond to the subject according to a magnitude relationship between a variation in position of the first region included in each classified group and a preset threshold value,
An imaging program for executing processing including calculating the in-focus position based on an average of phase differences detected by the phase difference detection unit included in the specified second region.

(Appendix 12)
When each variation in position of the first region included in the individual group is smaller than the preset threshold value, a group having a larger variation in position of the first region is selected from the individual groups, If at least one of the variations in the position of the first region included in the individual group is greater than the preset threshold, the group in which the variation in the position of the first region is smaller among the individual groups The imaging program according to claim 11, wherein the second region is specified from an average value and variation of positions of the first regions belonging to the selected group.

(Appendix 13)
A region within a range represented by a variation in the position of the first region belonging to the selected group, based on a point represented by an average value of the positions of the first region belonging to the selected group, The imaging program according to attachment 12, which is specified as the second region.

(Appendix 14)
The phase difference of each of the first regions is compared with the average value of the phase difference of each of the first regions, and each of the first regions is classified into the plurality of groups. The imaging program according to any one of the above.

(Appendix 15)
The imaging program according to any one of appendices 11 to 14, wherein the degree of variation in the position of the first region is expressed by a standard deviation.

(Appendix 16)
On the computer,
Through an optical system that can move the focal position by image sensors with multiple phase difference detectors that detect the phase difference for calculating the in-focus position using the image plane phase difference detection method. Image a subject represented by light incident on a light receiving surface of the image sensor;
A plurality of first regions set so as to include the phase difference detection units different from each other are grouped into a plurality of groups based on the magnitude relationship of the phase differences detected by the phase difference detection units included in the individual first regions. And specifying a second region that is estimated to correspond to the subject according to a magnitude relationship between a variation in position of the first region included in each classified group and a preset threshold value,
Computer-readable recording of an imaging program for executing processing including calculating the in-focus position based on an average of phase differences detected by the phase difference detection unit included in the specified second region Recording medium.

DESCRIPTION OF SYMBOLS 10 ... Imaging device, 12 ... Lens, 14L, 14R ... Micro lens, 16, 17, 18 ... Photoelectric conversion cell, 19 ... Imaging element, 20 ... Imaging part, 22 ... Distance measuring area, 24 ... Estimation window, 27 ... Closed Area 28... Main subject area 30. Setting section 32. Ranging area setting section 34. Phase difference calculation area setting section 40. Identification section 42. Phase difference acquisition section 44. Classification section 46. , 48 ... main subject region specifying part, 50 ... calculating part, 60 ... driving part, 70 ... contrast AF part, 100 ... computer, 102 ... CPU, 104 ... memory, 106 ... storage part, 108 ... bus, 112 ... input Device 114, display device 116, motor 120, imaging program 150, recording medium

Claims (7)

The phase difference detection unit for detecting the phase difference for calculating the in-focus position by the image plane phase difference detection method includes imaging elements arranged at a plurality of positions in the light receiving surface, and through an optical system that can move the focal position. An imaging unit that images a subject with light incident on a light receiving surface of the imaging element;
A plurality of first regions set so as to include the phase difference detection units different from each other are grouped into a plurality of groups based on the magnitude relationship of the phase differences detected by the phase difference detection units included in the individual first regions. And a specifying unit that specifies a second region estimated to correspond to the subject according to a magnitude relationship between a variation in position of the first region included in each classified group and a preset threshold value When,
A calculating unit that calculates the in-focus position based on an average of phase differences detected by the phase difference detecting unit included in the second region specified by the specifying unit;
An imaging apparatus including: In the case where each variation in the position of the first region included in the individual group is smaller than the preset threshold value, the specifying unit has a larger variation in the position of the first region in the individual group. When at least one of the variations in the position of the first region included in the individual group is larger than the preset threshold, the position of the first region in the individual group is selected. The imaging apparatus according to claim 1, wherein a group having a smaller variation is selected, and the second region is identified from an average value and a variation of each position of the first region belonging to the selected group. The specifying unit is within a range represented by a variation in the position of the first region belonging to the selected group, based on a point represented by an average value of the position of the first region belonging to the selected group. The imaging device according to claim 2, wherein the region is specified as the second region. The specifying unit compares each phase difference of the first region with an average value of each phase difference of the first region, and classifies each of the first regions into the plurality of groups. The imaging device according to any one of claims 1 to 3. The imaging device according to any one of claims 1 to 4, wherein a degree of variation in the position of the first region is expressed by a standard deviation. On the computer,
Through an optical system that can move the focal position by image sensors with multiple phase difference detectors that detect the phase difference for calculating the in-focus position using the image plane phase difference detection method. Image a subject represented by light incident on a light receiving surface of the image sensor;
A plurality of first regions set so as to include the phase difference detection units different from each other are grouped into a plurality of groups based on the magnitude relationship of the phase differences detected by the phase difference detection units included in the individual first regions. And specifying a second region that is estimated to correspond to the subject according to a magnitude relationship between a variation in position of the first region included in each classified group and a preset threshold value,
An imaging method for executing a process including calculating the in-focus position based on an average of phase differences detected by the phase difference detection unit included in the identified second region. On the computer,
Through an optical system that can move the focal position by image sensors with multiple phase difference detectors that detect the phase difference for calculating the in-focus position using the image plane phase difference detection method. Image a subject represented by light incident on a light receiving surface of the image sensor;
A plurality of first regions set so as to include the phase difference detection units different from each other are grouped into a plurality of groups based on the magnitude relationship of the phase differences detected by the phase difference detection units included in the individual first regions. And specifying a second region that is estimated to correspond to the subject according to a magnitude relationship between a variation in position of the first region included in each classified group and a preset threshold value,
An imaging program for executing processing including calculating the in-focus position based on an average of phase differences detected by the phase difference detection unit included in the specified second region.

Images