JP5375420B2 - Image processing apparatus, image processing method, and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To appropriately detect an image area which is conjectured that observers take attention from an image with little change. <P>SOLUTION: An imaging device 100 includes: a grouping unit 5c which carries out grouping of a plurality of pixels of acquired processing object images into a plurality of groups based on values of variables in a predetermined color space of respective pixels; a variance value calculation unit 5d which calculates a variance value in the image of pixels that belong to respective groups, for every groups of the plurality of groups carried out grouping; an evaluation value calculation unit 5g which calculates evaluation values of respective groups based on the calculated evaluation values; and an attention area detection unit 5h which detects an area composed of pixels that belong to any one group out of the plurality of the groups as the attention area based on the calculation result of the evaluation values of the respective groups. <P>COPYRIGHT: (C)2011,JPO&amp;INPIT

Description

  The present invention relates to an image processing apparatus, an image processing method, and a program for analyzing an image, for example.

  Conventionally, as a technique for detecting a region of interest from an image, it is speculated that a specific image region generated by the movement of a subject is detected from a difference value between two consecutive frames of the image, and that the region will receive attention. A technique that is regarded as a region is known (see, for example, Patent Document 1).

JP-A-9-205575

  However, since the technique described in Patent Document 1 is considered on the premise that “the observer pays attention to a moving subject in a moving image”, for example, from an image with little change such as a single still image. There is a problem that the region cannot be detected.

  Accordingly, an object of the present invention is to enable appropriate detection of an image area that is estimated to be observed by an observer from an image with little change.

In order to solve the above problem, an image processing apparatus according to claim 1 is provided.
An acquisition unit that acquires an image, a grouping unit that groups a plurality of pixels of the image acquired by the acquisition unit into a plurality of groups based on a value of a variable in a predetermined color space of each pixel, and the group For each of a plurality of groups grouped by the dividing unit, a variance value calculating unit that calculates a variance value in the image of pixels belonging to each group for each group, and based on the variance value calculated by the variance value calculating unit Then, based on the evaluation value calculation means for calculating the evaluation value of each group and the calculation result by the evaluation value calculation means, attention is paid to a region constituted by pixels belonging to any one of the plurality of groups. the ratio of the detection means for detecting a region, the new region of interest is detected by the detection means, and the detection result of a past predetermined number Was carried out, and determination means for determining whether to adopt the newly detected attention area according to the comparison result, characterized in that it comprises a.

An image processing apparatus according to a second aspect of the present invention provides :
An acquisition unit that acquires an image, a grouping unit that groups a plurality of pixels of the image acquired by the acquisition unit into a plurality of groups based on a value of a variable in a predetermined color space of each pixel, and the group For each of a plurality of groups grouped by the dividing unit, a variance value calculating unit that calculates a variance value in the image of pixels belonging to each group for each group, and based on the variance value calculated by the variance value calculating unit Then, based on the evaluation value calculation means for calculating the evaluation value of each group and the calculation result by the evaluation value calculation means, attention is paid to a region constituted by pixels belonging to any one of the plurality of groups. A detection means for detecting the area, and a predetermined area when the attention area detected by the detection means is composed of a plurality of closed areas. It is characterized in Rukoto and a exclude excluding means from said region of interest a small closed area than the size.

An image processing apparatus according to a third aspect of the present invention provides:
An acquisition unit that acquires an image, a grouping unit that groups a plurality of pixels of the image acquired by the acquisition unit into a plurality of groups based on a value of a variable in a predetermined color space of each pixel, and the group for each of Ri by the interface unit grouped plurality of groups, and the variance value calculating means for calculating a dispersion value in the image of the pixels belonging to each group for each group, the variance value calculated by the variance value calculating means Based on the evaluation value calculation means for calculating the evaluation value of each group, and an area configured by pixels belonging to any one of the plurality of groups based on the calculation result by the evaluation value calculation means Bei example a detector for detecting a region of interest, and the each of the plurality of groups, put on the predetermined color space defining each group The Rukoto such overlap the group adjacent ranges of the components are characterized.

An image processing apparatus according to a fourth aspect of the present invention provides :
An acquisition unit that acquires an image, a grouping unit that groups a plurality of pixels of the image acquired by the acquisition unit into a plurality of groups based on a value of a variable in a predetermined color space of each pixel, and the group For each of the plurality of groups grouped by the dividing means, a centroid calculating means for calculating the centroid position of the pixels belonging to each group for each group, and the centroid position calculated by the centroid calculating means and the center of the image Based on the distance calculation means for calculating the distance, the evaluation value calculation means for calculating the evaluation value of each group based on the distance calculated by the distance calculation means, and the calculation result by the evaluation value calculation means, Detecting means for detecting, as a region of interest, an area formed by pixels belonging to any one of a plurality of groups; When a detection area is newly detected by the detection means, a comparison is made with a predetermined number of past detection results, and a decision is made to determine whether to adopt the newly detected attention area according to the comparison result and means, the is characterized in Rukoto.

An image processing apparatus according to a fifth aspect of the present invention provides :
An acquisition unit that acquires an image, a grouping unit that groups a plurality of pixels of the image acquired by the acquisition unit into a plurality of groups based on a value of a variable in a predetermined color space of each pixel, and the group For each of the plurality of groups grouped by the dividing means, a centroid calculating means for calculating the centroid position of the pixels belonging to each group for each group, and the centroid position calculated by the centroid calculating means and the center of the image Based on the distance calculation means for calculating the distance, the evaluation value calculation means for calculating the evaluation value of each group based on the distance calculated by the distance calculation means, and the calculation result by the evaluation value calculation means, Detecting means for detecting, as a region of interest, an area formed by pixels belonging to any one of a plurality of groups; When the detected region of interest by the detection means is composed of a plurality of the closed region, it is characterized by comprising, Excluding excluding means from said region of interest a small closed region than the predetermined size.

An image processing apparatus according to a sixth aspect of the present invention provides :
An acquisition unit that acquires an image, a grouping unit that groups a plurality of pixels of the image acquired by the acquisition unit into a plurality of groups based on a value of a variable in a predetermined color space of each pixel, and the group For each of the plurality of groups grouped by the dividing means, a centroid calculating means for calculating the centroid position of the pixels belonging to each group for each group, and the centroid position calculated by the centroid calculating means and the center of the image Based on the distance calculation means for calculating the distance, the evaluation value calculation means for calculating the evaluation value of each group based on the distance calculated by the distance calculation means, and the calculation result by the evaluation value calculation means, Detecting means for detecting, as a region of interest, an area formed by pixels belonging to any one of a plurality of groups; For example, each of the plurality of groups is characterized by comprising overlapping said predetermined group of adjacent ranges of each component in the color space defining each group.

The image processing method of the invention according to claim 7 is:
An acquisition step of acquiring an image, a grouping step of grouping a plurality of pixels of the image acquired in the acquisition step into a plurality of groups based on a value of a variable in a predetermined color space of each pixel; For each of the plurality of groups grouped in the grouping step, a variance value calculating step for calculating the variance value in the image of the pixels belonging to each group for each group, and the variance calculated in this variance value calculating step based on the value, the evaluation value calculation step of calculating an evaluation value of each group, based on the definitive calculation result to the evaluation value calculating step is configured by any pixels that belong to one group of the plurality of groups a detection step of detecting that region as a region of interest, a new region of interest is detected at the detection step , Compares the detection result of past predetermined number of times, it is characterized by executing a determining step of determining whether to adopt the newly detected attention area according to the comparison result.

An image processing method according to an eighth aspect of the present invention includes:
An acquisition step of acquiring an image, a grouping step of grouping a plurality of pixels of the image acquired in the acquisition step into a plurality of groups based on a value of a variable in a predetermined color space of each pixel; for each of a plurality of groups grouped by the grouping step, the variance value calculation step of calculating a variance value for each group in the image of the pixels belonging to each group was calculated dispersed by the dispersion value calculation step An evaluation value calculating step for calculating an evaluation value of each group based on the value , and a pixel belonging to any one of the plurality of groups based on a calculation result in the evaluation value calculating step an image processing method for performing a detection step of detecting a region as a region of interest, wherein the plurality of guru Each flop is characterized by comprising overlapping said predetermined group of adjacent ranges of each component in the color space defining each group.

The program of the invention according to claim 9 is:
An acquisition means for acquiring an image, a grouping means for grouping a plurality of pixels of the image acquired by the acquisition means into a plurality of groups based on values of variables in a predetermined color space of each pixel, for each of a plurality of groups grouped I by the grouping means, the center of gravity calculating means for calculating the center of gravity of the pixels belonging to each group for each group, of the image and the calculated barycentric position by the centroid calculating unit distance calculating means for calculating the distance between the center, based on the distance calculated by the distance calculating means, evaluation value calculating means for calculating an evaluation value of each group, based on the calculation result by the evaluation value calculating means, Detection means for detecting an area formed by pixels belonging to any one of the plurality of groups as an attention area When the detected region of interest by the detection means is composed of a plurality of the closed region, it is characterized in that to function as excluding excluding means, from the region of interest smaller closed region than the predetermined size.

The program of the invention according to claim 10 is:
An acquisition means for acquiring an image, a grouping means for grouping a plurality of pixels of the image acquired by the acquisition means into a plurality of groups based on values of variables in a predetermined color space of each pixel, For each of the plurality of groups grouped by the grouping means, the centroid calculating means for calculating the centroid position of the pixels belonging to each group for each group, the centroid position calculated by the centroid calculating means and the center of the image A distance calculation means for calculating a distance, an evaluation value calculation means for calculating an evaluation value of each group based on the distance calculated by the distance calculation means, and a plurality of the plurality of evaluation values based on a calculation result by the evaluation value calculation means Detection means for detecting, as a region of interest, an area formed by pixels belonging to any one of the groups A program to function as each of said plurality of groups is characterized by comprising overlapping said predetermined group of adjacent ranges of each component in the color space defining each group.

  According to the present invention, it is possible to appropriately detect an image region that is estimated to be observed by an observer from an image with little change.

It is a block diagram which shows schematic structure of the imaging device of one Embodiment to which this invention is applied. 3 is a flowchart illustrating an example of an operation related to an attention area output process by the imaging apparatus of FIG. 1. It is a flowchart which shows an example of the operation | movement which concerns on the attention area detection process in attention area output processing. It is a figure for demonstrating the attention area output by attention area output processing. It is a flowchart which shows an example of the operation | movement which concerns on the attention area stabilization process in attention area output processing.

Hereinafter, specific embodiments of the present invention will be described with reference to the drawings. However, the scope of the invention is not limited to the illustrated examples.
FIG. 1 is a block diagram illustrating a schematic configuration of an imaging apparatus 100 according to an embodiment to which the present invention is applied.

The imaging apparatus 100 according to the present embodiment divides a plurality of pixels of a processing target image into a plurality of groups based on values of variables in a predetermined color space of each pixel, and pixels belonging to each of the plurality of groups An area composed of pixels belonging to any one of a plurality of groups, based on the evaluation value of each group, based on the evaluation value of each group based on the variance value in the image Are detected as the attention area A1. Here, the attention area is an area that is estimated to be noticed when the photographer (observer) views the processing target image, and is output by attention area output processing described later.
Specifically, as illustrated in FIG. 1, the imaging device 100 includes an imaging unit 1, an imaging control unit 2, an image data generation unit 3, a memory 4, an image processing unit 5, and a display control unit 6. , A display unit 7, a recording medium 8, an operation input unit 9, and a central control unit 10.

  The imaging unit 1 captures a subject and generates an image frame. Specifically, although not shown, the imaging unit 1 includes a lens unit including a plurality of lenses such as a zoom lens and a focus lens, a diaphragm that adjusts the amount of light passing through the lens unit, and a CCD ( The image sensor includes an electronic image pickup unit that includes an image sensor such as a charge coupled device (CMOS) or a complementary metal-oxide semiconductor (CMOS), and converts an optical image that has passed through various lenses of the lens unit into a two-dimensional image signal.

Although not shown, the imaging control unit 2 includes a timing generator, a driver, and the like. Then, the imaging control unit 2 scans and drives the electronic imaging unit with a timing generator and a driver, converts the optical image into a two-dimensional image signal with the electronic imaging unit every predetermined period, and captures the imaging area of the electronic imaging unit Image frames are read out for each screen and output to the image data generation unit 3.
The imaging control unit 2 performs AE (automatic exposure processing), AF (automatic focusing processing), AWB (automatic white balance), and the like as adjustment control of imaging conditions.

The image data generation unit 3 appropriately adjusts the gain for each of the RGB color components with respect to the analog value signal of the image data transferred from the electronic imaging unit, and then samples and holds the sample by a sample hold circuit (not shown). The digital data is converted into digital data by a D converter (not shown), and color processing including pixel interpolation processing and γ correction processing is performed by a color process circuit (not shown), and then a luminance signal Y and a color difference signal Cb, Cr (image data in YUV color space) is generated.
The luminance signal Y and the color difference signals Cb and Cr output from the color process circuit are DMA-transferred to a memory 4 used as a buffer memory via a DMA controller (not shown).

  The memory 4 is constituted by, for example, a DRAM or the like, and temporarily stores data processed by the image processing unit 5, the central control unit 10, and the like.

The image processing unit 5 includes an image acquisition unit (acquisition unit) 5a that acquires an image captured by the imaging unit 1 as a processing target image.
Specifically, the image acquisition unit 5a acquires YUV data of an image generated by the image data generation unit 3 at the time of capturing a live view image or a recording image, for example.

Further, the image processing unit 5 includes a smoothing unit (smoothing means) 5b for performing a smoothing process for smoothing the color of the processing target image on the processing target image acquired by the image acquisition unit 5a. Yes.
The smoothing process is a process performed to smooth a fine color distribution in a processing target image and catch a rough color change. That is, the smoothing unit 5b smoothes the color of the processing target image using a predetermined filter such as a Gaussian filter, an averaging filter, or an ε filter, for example.

Further, the image processing unit 5 is a grouping unit that groups a plurality of pixels of the processing target image subjected to the smoothing process by the smoothing unit into a plurality of groups defined by variable ranges in a predetermined color space. 5c.
Specifically, the grouping unit 5c preliminarily describes each component of a predetermined color space (for example, HSV color space) (for example, in the case of the HSV color space, hue, saturation, and lightness (Value)). ) Is divided into a plurality of groups according to a predetermined setting standard. At this time, the grouping unit 5c may set a plurality of groups such that a range defining one group overlaps a range defining adjacent groups for each component variable in a predetermined color space. Accordingly, it is possible to eliminate omission of grouping for pixels existing in the boundary region between groups, and it is possible to easily group into a plurality of groups in consideration of color ambiguity.

For example, when the HSV color space is applied as the predetermined color space, the grouping unit 5c sets the value of 0 ° to 360 ° for the hue (H) as 0 ° to 60 ° as the group H0, and 30 ° ~ 90 ° is Group H1, 60 °-120 ° is Group H2, 90 °-150 ° is Group H3, 120 °-180 ° is Group H4, 150 °-210 ° is Group H5, 180 ° ~ 240 ° as group H6, 210 °-270 ° as group H7, 240 °-300 ° as group H8, 270 °-330 ° as group H9, 300 °-360 (0) ° as group H10 Twelve groups are set by setting 330 ° to 30 ° as group H11.
Also, the grouping unit 5c assumes that the value of saturation (S) is 0 to 255, 0 to 127 is group S0, 64 to 159 is group S1, 128 to 223 is group S2, and 160 to 255. By setting the group to S3, four groups are set.
Further, the grouping unit 5c assumes that the value (b) is 0 to 255, 0 to 127 is the group V0, 64 to 159 is the group V1, 128 to 223 is the group V2, and 160 to 255 is the group V0. Four groups are set by setting to group V3.
When the hue (H), saturation (S), and brightness (V) are integrated, a total of 192 groups are set.
The group setting of each component of the HSV color space described above is an example, and is not limited to these. Further, the HSV color space is exemplified as the predetermined color space because it is close to human color perception and easy to grasp intuitively. However, the color space is not limited to this, and the RGB color space or CIEL * a * b is used depending on the application. * Other color spaces such as a color space may be applied.

Then, the grouping unit 5c converts the image data in the YUV color space that has been smoothed by the smoothing unit 5b into image data in the HSV color space according to a predetermined conversion formula, and then the image data in the HSV color space. Are grouped into a plurality of groups based on the values of the variables in the HSV color space of each pixel. In other words, the grouping unit 5c calculates 192 values for each of the plurality of pixels of the image data in the HSV color space from the values (variable values) of the components of hue, saturation, and lightness (Value). It identifies and identifies which group it belongs to.
For example, in the image data of the HSV color space, pixels having values of hue (H): 45 °, saturation (S): 90, lightness (V): 200 belong to groups H0 and H1 for hue. The saturation belongs to the groups S0 and S1, and the brightness belongs to the groups V2 and V3. Therefore, when the hue, saturation, and lightness are integrated, the pixel becomes (H0, S0, V2), (H1, S0, V2), (H0, S1, V2), (H1, S1, V2), (H0 , S0, V3), (H1, S0, V3), (H0, S1, V3), and (H1, S1, V3).

  Here, the grouping unit 5c constitutes grouping means for grouping a plurality of pixels of the image acquired by the image acquisition unit 5a into a plurality of groups based on the values of variables in the predetermined color space of each pixel. doing.

In addition, the image processing unit 5 calculates, for each of the plurality of groups grouped by the grouping unit 5c, a variance value calculation unit (dispersion value calculation unit) that calculates a variance value in a processing target image of a plurality of pixels belonging to each group. ) 5d.
For example, the variance value calculation unit 5d calculates, for each group, variance values in the processing target image of a plurality of pixels belonging to each group. The calculation of the variance value by the variance value calculation unit 5d is performed according to a known arithmetic expression, and a detailed description thereof is omitted.
In addition, the number of pixels may be counted for each group, and a group having the number of pixels equal to or less than a predetermined threshold or greater than or equal to a predetermined threshold may be excluded from the processing for calculating the variance value.

Further, the image processing unit 5 calculates, for each of a plurality of groups grouped by the grouping unit 5c, a centroid calculation unit (centroid calculation means) 5e that calculates a centroid position in a processing target image of a plurality of pixels belonging to each group. It has. The calculation of the center of gravity position by the center of gravity calculation unit 5e is performed according to a known arithmetic expression, and detailed description thereof is omitted.
In addition, the number of pixels may be counted for each group, and a group having the number of pixels equal to or less than a predetermined threshold or greater than or equal to a predetermined threshold may be excluded from the process of calculating the center of gravity.

The image processing unit 5 includes a distance calculation unit (distance calculation unit) 5f that calculates the distance between the center of gravity of each group calculated by the center of gravity calculation unit 5e and the center of the processing target image.
In other words, since a general camera user tends to place the region of interest at the center of the image, the distance between the center of gravity of each group and the center of the processing target image is considered when calculating the evaluation value described later. Thus, the distance calculation unit 5f calculates the distance D from the center of the processing target image to the centroid position of each group.

In addition, the image processing unit 5 includes an evaluation value calculation unit 5g that calculates the evaluation values of each of the plurality of groups.
The evaluation value calculation unit 5g calculates the evaluation value of each group based on the variance value calculated by the variance value calculation unit 5d and the distance D calculated by the distance calculation unit 5f for each group. Specifically, the evaluation value calculation unit 5g stores, for example, a weighting table in which the evaluation value increases as the distance D increases, and a value obtained by multiplying the variance value by the weighting value according to the distance D is stored. An evaluation value or a value calculated according to a predetermined arithmetic expression defined by the following formula may be used as the evaluation value.
Evaluation value = dispersion value × (a * D ^ 2 + b)
(A, b; predetermined constants)

  Here, the evaluation value calculator 5g constitutes an evaluation value calculator that calculates the evaluation value of each group based on the variance value calculated by the variance value calculator 5d.

Further, the image processing unit 5 includes an attention region detection unit 5h that detects the attention region A1 from the processing target image based on the calculation result by the evaluation value calculation unit 5g.
The attention area detection unit 5h identifies any one group having the smallest evaluation value among a plurality of groups based on the calculation result of the evaluation values of all the processing target groups by the evaluation value calculation unit 5g, An area formed by pixels belonging to the group is detected as an attention area A1 (see FIG. 4A). In other words, the evaluation value calculated by the evaluation value calculation unit 5g does not exist as a group of specific pixels belonging to a group expressed in a predetermined color space in the processing target image as the variance value increases (that is, As the distance D increases (that is, the specific pixel belonging to the group expressed in the predetermined color space is further away from the center of the processing target image), the value increases. Therefore, the attention area detection unit 5h is a color in which pixels belonging to the group having the smallest evaluation value in consideration of the variance value and the distance D are arranged in the processing target image, and the center of the processing target image Is determined to be located at a short distance, and an area composed of pixels belonging to the group having the smallest evaluation value is detected as the attention area A1.
Here, the attention area detection unit 5h detects, as the attention area A1, an area composed of pixels belonging to any one of the plurality of groups based on the calculation result by the evaluation value calculation unit 5g. Is configured.

The image processing unit 5 compares the attention area A1 detected by the attention area detection section 5h with the newly detected attention area A1, and the newly detected attention area A1 is the detected attention. A comparison / determination unit (determination unit) 5i that determines whether or not the area has changed from the area A1 is provided.
Specifically, in the attention area stabilization process (see FIG. 5) described later, the comparison determination unit 5i includes a plurality of attention areas A1 detected in the past stored in the memory 4 and the detected attention. The attention area A1 newly detected after the detection of the area A1 is compared. Then, the comparison determination unit 5i determines whether or not the newly detected attention area A1 has changed to a different area from the attention area A1 detected in the past, according to the comparison result.
The comparison determination between the newly detected attention area A1 and the detected attention area A1 includes, for example, a group name that defines a group to which the pixels constituting each attention area A1 belong, the number of pixels in the group, the group This is performed using the center of gravity position.

Further, the image processing unit 5 controls the output content so that the output content of the attention region A1 (main attention region A2) detected by the attention region detection unit 5h is stabilized according to the determination result by the comparison determination unit 5i. An output control unit (output control means) 5j that performs attention area stabilization processing is provided.
That is, for example, when a live view image or an image at the time of recording is updated and displayed on the display unit 7, the output process of the attention area A1 is periodically performed, and the output result of the attention area A1 or the attention area A1 is displayed. You may make it display continuously the figure (for example, the minimum rectangular frame etc. which surround attention area A1) etc. expressed approximately. In this case, there is a possibility that the output result of the attention area A1 or the display of the approximate graphic will change excessively and the user will be hard to see. Therefore, the output control unit 5j determines the result of determination between the attention area A1 newly detected by the comparison determination section 5i and the detected attention area A1, for example, a group that defines the group to which the pixels constituting each attention area A1 belong. The attention is stabilized so that the output contents of the attention area A1 output from the attention area detection unit 5h are not excessively changed according to the change contents such as the name, the number of pixels of the group, and the center of gravity position of the group. Perform region stabilization processing. Specifically, for example, in the attention area stabilization process, for the latest main attention area and the current main attention area corresponding to “current group name”, the centroid position of the group to which the pixels constituting each attention area belong Alternatively, when it is determined that the number of pixels in the group has not changed by a predetermined ratio or more, the output control unit 5j performs a process of averaging the current main attention area and the latest main attention area.

The display control unit 6 performs control for reading display image data temporarily stored in the memory 4 and displaying the read image data on the display unit 7.
Specifically, the display control unit 6 includes a VRAM, a VRAM controller, a digital video encoder, and the like (all not shown). The digital video encoder periodically reads the luminance signal Y and the color difference signals Cb and Cr read from the memory 4 and stored in the VRAM under the control of the central control unit 10 from the VRAM via the VRAM controller. It reads out, generates a video signal based on these data, and outputs it to the display unit 7.

  The display unit 7 is, for example, a liquid crystal display device, and displays an image captured by the electronic imaging unit based on a video signal from the display control unit 6 on a display screen. Specifically, the display unit 7 updates the live view image while sequentially updating a plurality of image frames generated by imaging of the subject by the imaging unit 1 at a predetermined frame rate in the still image capturing mode and the moving image capturing mode. indicate. The display unit 7 displays an image (rec view image) recorded as a still image or an image being recorded as a moving image.

  The recording medium 8 is composed of, for example, a non-volatile memory (flash memory) or the like, and recording still image data encoded in a predetermined compression format by an encoding unit (not shown) of the image processing unit 5 or a plurality of recording image data The moving image data composed of image frames is stored.

  The operation input unit 9 is for performing a predetermined operation of the imaging apparatus 100. Specifically, the operation input unit 9 includes a shutter button according to a subject photographing instruction, a selection determination button according to a selection instruction for an imaging mode, a function, and the like, a zoom button according to a zoom amount adjustment instruction, and the like (all illustrated). Abbreviation), a predetermined operation signal is output to the central control unit 10 in accordance with the operation of these buttons.

  The central control unit 10 controls each unit of the imaging device 100. Specifically, the central control unit 10 includes a CPU (not shown) and performs various control operations in accordance with various processing programs (not shown) for the imaging apparatus 100.

Next, attention area output processing when the imaging device 100 captures a subject will be described with reference to FIGS.
FIG. 2 is a flowchart illustrating an example of an operation related to the attention area output process.

  As shown in FIG. 2, the image acquisition unit 5 a of the image processing unit 5 captures a YUV of a processing target image generated by the image data generation unit 3 by, for example, capturing a live view image or a recording image by the imaging unit 1. Color space image data is acquired (step S1). Subsequently, the smoothing unit 5b of the image processing unit 5 performs a smoothing process for smoothing the color of the processing target image using a predetermined filter (for example, a Gaussian filter or the like) (step S2).

Next, the image processing unit 5 performs attention area detection processing for detecting the attention area A1 from the processing target image (step S3).
Here, the attention area detection processing will be described in detail with reference to FIG.

FIG. 3 is a flowchart illustrating an example of an operation related to the attention area detection process.
As shown in FIG. 3, the grouping unit 5c converts the image data in the YUV color space that has been smoothed by the smoothing unit 5b into image data in the HSV color space according to a predetermined conversion formula (step S31). . Thereafter, the grouping unit 5c, for each of the plurality of pixels of the image data in the HSV color space after the conversion, values (coordinates) of the variables of the respective components of hue (Hue), saturation (Saturation), and lightness (Value). From among a plurality of preset groups (for example, 192), it is identified and labeled which group belongs (step S32).

  Next, the image processing unit 5 excludes a group whose number of pixels is equal to or less than a predetermined threshold value or more than a predetermined threshold value from among a plurality of labeled groups, and then sets the plurality of groups to be processed. One of the processing target groups is designated from among the above (step S33).

  Then, the center-of-gravity calculation unit 5e of the image processing unit 5 calculates the center-of-gravity positions of a plurality of pixels belonging to the processing target group according to a predetermined arithmetic expression (step S34). Subsequently, the variance value calculation unit 5d of the image processing unit 5 performs, for example, the variance in the processing target image according to a predetermined arithmetic expression based on the hue, saturation, and brightness components of a plurality of pixels belonging to the processing target group. A value is calculated (step S35). Next, the distance calculation unit 5f of the image processing unit 5 calculates the distance D from the center of the processing target image to the center of gravity position of the processing target group (step S36).

  Thereafter, the evaluation value calculation unit 5g of the image processing unit 5 evaluates each group based on the variance value calculated by the variance value calculation unit 5d and the distance D calculated by the distance calculation unit 5f related to the processing target group. A value is calculated (step S37).

Next, the image processing unit 5 performs a process of determining whether or not the evaluation values of all the groups to be processed have been calculated (step S38). Here, if it is determined that the evaluation values of all the groups have not been calculated (step S38; NO), after specifying the next processing target group (step S39), the process proceeds to step S34, The subsequent processing is executed.
The processing after step S34 is repeatedly performed until it is determined in step S38 that the evaluation values of all the groups have been calculated (step S38; YES).

  If it is determined in step S38 that the evaluation values of all the groups have been calculated (step S38; YES), the attention area detection unit 5h of the image processing unit 5 performs all processing by the evaluation value calculation unit 5g. Based on the calculation result of the group evaluation value, one of the plurality of groups having the smallest evaluation value is specified (step S40), and an area constituted by pixels belonging to the group is set as the attention area A1. (See FIG. 4A) (step S41), and the attention area detection process is terminated.

Next, the image processing unit 5 returns to the process of FIG. 2, for example, an opening process, a process of labeling each closed area and erasing an area where the number of pixels does not satisfy a predetermined threshold, or a smoothing process. A graph cut process or the like that is repeatedly executed is performed to delete a small area in the attention area A1 and perform a shaping process that extracts only the main attention area A2 (see FIG. 4B) (step S4).
Subsequently, the image processing unit 5 performs a process of determining whether or not an instruction to execute the attention area stabilization process that stabilizes the output content of the main attention area A2 extracted by the shaping process is given (step S5). .
The execution instruction of the attention area stabilization process may be performed in advance based on a predetermined operation of the operation input unit 9 by the user, or may be set as a default according to the shooting mode.
When it is determined that the execution of the attention area stabilization process is not instructed (step S5; NO), the image processing unit 5 sets the main attention area A2 extracted by the shaping process as the final attention area A1. The data is output to the memory 4 (step S6).

On the other hand, if it is determined that the execution of the attention area stabilization process is instructed (step S5; YES), the image processing unit 5 performs the attention area stabilization process (step S7).
Here, the attention area stabilization processing will be described in detail with reference to FIG.

FIG. 5 is a flowchart illustrating an example of an operation related to the attention area stabilization process.
As shown in FIG. 5, the image processing unit 5 processes a comparison between the main attention area A2 detected by the attention area detection section 5h and the newly detected main attention area A2 in a loop (steps S71 to S75). ).
Specifically, the image processing unit 5 holds a plurality of (N) main attention areas A2 extracted by the shaping process, and acquires a group name, the number of pixels, and the barycentric position for each of them. Loop processing is performed so as to go back to the past in order from the newest one. First, the comparison / determination unit 5i of the image processing unit 5 calculates the group name (group name number i = 0) and the past k-th (k is an arbitrary real number from 1 to N) related to the latest main attention area. The group name (group name number i = k) of A2 is collated, and whether or not they match is stored for each loop count (step S72). Next, the comparison determination unit 5i determines whether or not the group name related to the latest main attention area is the same as the group name of the past main attention area A2 for a predetermined number of times (for example, three times) consecutively. Determination is made (step S73).

When it is determined in step S73 that the group name related to the latest main attention area is not the same as the group name of the past main attention area A2 for a predetermined number of consecutive times from the latest one (step S73; NO), the comparison is performed. The determination unit 5i determines whether or not the group name of the main attention area A2 is the same B times in A times in the past from the latest one (step S74). For example, the comparison / determination unit 5i determines whether or not the group name is the same four times in the past five times and whether or not the group name is the same five times in the past seven times. Note that the values of A and B can be arbitrarily changed according to the required stability. Further, the values of A and B may be changed and the determination may be made a plurality of times in one loop.
The above processing is repeated until the past Nth group whose group name number i is N (step S75).

When it is determined in step S74 that the group name of the main attention area A2 is not the same B times in the past A times from the latest one (step S74; NO), the image processing unit 5 The latest output result in the region A2 is regarded as a transient state. Then, after the image processing unit 5 updates and holds the latest main attention area as past data (step S76), the comparison determination section 5i includes the group names related to the latest main attention area in the latest C times. It is determined whether or not (step S77). Note that the value of C can be arbitrarily changed according to the required stability.
Here, if it is determined that the group name related to the latest main attention area is not included in the latest C times (step S77; NO), the image processing unit 5 determines that the latest main attention area is the main attention so far. It is considered that the area has changed to a completely different area from the area A2. Then, the image processing unit 5 updates and holds the group name related to the latest main attention area as the “current group name” that defines the current main attention area (step S78), and then outputs the output result of the latest main attention area. The final attention area is output to the memory 4 (step S79), and the attention area stabilization processing is terminated.

  On the other hand, when it is determined in step S77 that the group name related to the latest main attention area is included in the latest C times (step S77; YES), the image processing unit 5 outputs the result of the main attention area A2. Has not changed excessively and is now considered stable. Then, the image processing unit 5 outputs the current main attention area held as “current group name” as the final attention area to the memory 4 (step S80), and ends the attention area stabilization processing.

Whether or not it is determined in step S73 that the group name related to the latest main attention area is the same as the past group name of the main attention area A2 a predetermined number of times in succession (step S73; YES). Alternatively, when it is determined in step S74 that the group name of the main attention area A2 is the same as the Bth of the A times in the past from the latest one (step S74; YES), the image processing unit 5 The latest main attention area is updated and held as past data (step S81).
Subsequently, the comparison determination unit 5i of the image processing unit 5 determines whether or not the group name related to the latest main attention area is the same as the “current group name” that defines the current main attention area (step S82). ).

If it is determined in step S82 that the group name related to the latest main attention area is not the same as the “current group name” (step S82; NO), the image processing unit 5 determines that the attention area has been changed. Then, the process proceeds to step S78, and the group name related to the latest main attention area is updated and held as “current group name” (step S78), and the subsequent processes are executed.
On the other hand, when it is determined in step S82 that the group name related to the latest main attention area is the same as the “current group name” (step S82; YES ), the comparison determination unit 5i of the image processing unit 5 Determine whether or not the position of the center of gravity of the group to which the pixels making up each region of interest belong or the number of pixels in the group has changed by more than a predetermined ratio for the main region of interest of the current and the current main region of interest corresponding to the “current group name” (Step S83).
In the determination in step S83, not only the center of gravity position and the number of pixels of the main attention area A2, but also the position and size of the rectangular frame surrounding the main attention area A2, the distribution of the main attention area A2, etc. Any one that can be specified may be used.

If it is determined in step S83 that the position of the center of gravity of the group or the number of pixels in the group has changed by a predetermined ratio or more (step S83; YES), the image processing unit 5 has greatly changed the shape and position of the main attention area A2. Therefore, the process proceeds to step S78, and the group name related to the latest main attention area is updated and held as “current group name” (step S78), and the subsequent processes are executed.
On the other hand, if it is determined in step S83 that the position of the center of gravity of the group or the number of pixels in the group has not changed by a predetermined ratio or more (step S83; YES), the image processing unit 5 corresponds to “current group name”. An average of the current main attention area and the latest main attention area is output to the memory 4 as a final attention area (step S84), and the attention area stabilization processing is terminated.

  As described above, according to the imaging apparatus 100 of the present embodiment, each of a plurality of groups in which a plurality of pixels of a processing target image are grouped based on a distribution in a predetermined color space (for example, HSV color space). , The evaluation value of each group can be calculated based on the variance value in the image of at least one pixel belonging to each group, and the attention area A1 can be detected based on the calculation result of the evaluation value of each group. That is, based on the distribution of pixels in a predetermined color space, an area composed of pixels of a color that are present in a processing target image can be detected as the attention area A1, and therefore, fewer operations than in the past. The attention area A1 can be detected by the amount, and the detection processing of the attention area A1 can be speeded up even with the imaging device 100 having a low calculation capability.

  The evaluation of each group is performed based on the variance value calculated by the variance value calculation unit 5d and the distance between the center of gravity of each group calculated by the distance calculation unit 5f and the center of the image for each of the plurality of groups. Since the value can be calculated, the distribution of a plurality of pixels in the processing target image can be taken into account when calculating the evaluation value of each group, and the processing target is displayed in a color that exists in the processing target image. An area formed by pixels closer to the center of the image can be detected as the attention area A1. Thereby, the detection accuracy of attention area A1 can be improved more.

  Further, depending on the result of determining whether or not the main attention area A2 newly detected after the detection of the main attention area A2 (attention area A1) has changed to a different area from the detected main attention area A2, the main attention area Since the attention area stabilization process for controlling the output contents so as to stabilize the output contents of the area A2 is performed, the output contents of the main attention area A2 can be prevented from changing excessively. For example, the imaging apparatus 100 Even if the direction of the screen slightly changes, the display unit 7 can stably display the detection result of the main attention area A2 and approximate figures.

The present invention is not limited to the above-described embodiment, and various improvements and design changes may be made without departing from the spirit of the present invention.
For example, in the above embodiment, as a result of calculating the evaluation value of each group, the region having the lowest evaluation value is detected as the attention region A1. An area where the evaluation value is equal to or less than a predetermined value may be detected as the attention area A1, and an area existing in the vicinity of the detected area and a predetermined color space (for example, HSV space) is also set as the attention area. Alternatively, an area composed of pixels in the vicinity of the color space and in the vicinity of the actual image may be set as the attention area.
In addition, after the smoothing process (step S2) by the smoothing unit 5b, a temporary attention area is estimated by excluding an imaging target with a low degree of attention of the user located closer to the end in the composition (view angle). Processing may be performed. That is, by tracking a pixel similar to the at least one pixel in the peripheral area of at least one pixel at the four corners of the processing target image after the smoothing process, the end side in the composition (view angle) An imaging target with a low degree of attention of the user located in the image is identified, and a region other than the region with a low degree of attention of the user configured by the tracked pixels in the processing target image is estimated as a temporary attention region good. Then, by subsequently detecting the attention area A1 as a true attention area by the above processing, the detection accuracy of the attention area A1 from the processing target image can be further improved. Further, among each group, a group including pixels located on the edge side of the processing target image may not calculate the evaluation value, or includes many pixels located on the edge side of the processing target image. As the evaluation value increases, weighting may be performed. That is, the group closer to the center of the image may be more easily output as the attention area.

In the above embodiment, the evaluation value of each group is calculated based on the variance value calculated by the variance value calculation unit 5d and the distance D calculated by the distance calculation unit 5f. The value calculation method is not limited to this. For example, one of the variance value calculated by the variance value calculation unit 5d and the distance D calculated by the distance calculation unit 5f is calculated as the evaluation value of each group. May be.
That is, in calculating the evaluation value of each group by calculating the distance D calculated by the distance calculation unit 5f as the evaluation value of each group in the attention area output processing (see FIG. 2). Since the distribution of a plurality of pixels in the target image can be taken into account, and an area formed by pixels closer to the center of the processing target image can be detected as the attention area A1, attention is paid with a smaller amount of computation than in the past. The area A1 can be detected, and even the imaging device 100 with low computing capability can speed up the process of detecting the attention area A1.

  The attention area generally tends to be in the center of the image. Therefore, the evaluation of each group is performed based on the variance value calculated by the variance value calculation unit 5d and the distance between the center of gravity of each group calculated by the distance calculation unit 5f and the center of the image for each of the plurality of groups. A value can be calculated, and a group closer to the center of the image can be more easily output as a region of interest.

Further, in the above embodiment, in the attention area output process (see FIG. 2), the smoothing process (step S2) is performed on the process target image acquired by the image acquisition unit 5a. Whether or not to perform the smoothing process can be arbitrarily changed as appropriate.
Further, after the attention area detection process (step S3), a shaping process (step S4) is performed in which a small area in the attention area A1 is deleted and only the main attention area A2 (see FIG. 4B) is extracted. However, whether or not to perform the shaping process can be arbitrarily changed as appropriate. The shaping process may be performed after the attention area stabilization process (step S7) or the attention area output (step S6).

  Furthermore, in the above-described embodiment, the attention content is controlled so that the output content of the main attention area A2 extracted by the shaping process (step S4) is stabilized in the attention area output processing (see FIG. 2). Although the region stabilization processing (step S7) is performed, whether or not to perform the region of interest stabilization processing can be arbitrarily changed as appropriate. In addition, after the attention area stabilization process is performed, the output attention area may be averaged with the past n output areas. That is, the attention area that has been held for the past n times and the attention area output this time may be mixed at n: 1.

In addition, the configuration of the imaging apparatus 100 is merely an example illustrated in the above embodiment, and is not limited thereto. That is, the imaging apparatus 100 is exemplified as the image processing apparatus, but is not limited thereto. For example, the processing target image is captured by an imaging device different from the imaging device 100, the image data of the processing target image transferred from the imaging device is acquired, and the attention area A1 is detected. It may be.
Further, when a live view image or an image at the time of recording is updated and displayed on the display unit 7, the output process of the attention area A1 is periodically performed, and the output result of the attention area A1 or the attention area A1 is approximated. (For example, the smallest rectangular frame surrounding the attention area A1) or the like may be continuously displayed.
Furthermore, for example, when the photographer instructs the imaging apparatus 100 to perform photographing by a predetermined operation while outputting the attention area A1 extracted from the live view image, AF is performed based on the output attention area A1. (Autofocus) or AE (automatic exposure) processing may be performed.

  In addition, in the above embodiment, the functions as an acquisition unit, a grouping unit, a variance value calculation unit, a center of gravity calculation unit, a distance calculation unit, an evaluation value calculation unit, and a detection unit are controlled under the control of the central control unit 10. Then, the image acquisition unit 5a, grouping unit 5c, variance value calculation unit 5d, centroid calculation unit 5e, distance calculation unit 5f, evaluation value calculation unit 5g, and attention area detection unit 5h of the image processing unit 5 are driven. However, the present invention is not limited to this, and may be realized by executing a predetermined program or the like by the CPU of the central control unit 10.

That is, a program including an acquisition process routine, a grouping process routine, a variance value calculation process routine, an evaluation value calculation process routine, and a detection process routine is stored in a program memory (not shown) that stores the program. Then, the CPU of the central control unit 10 may function as an acquisition unit that acquires an image by an acquisition process routine. Further, the CPU of the central control unit 10 as a grouping means for grouping a plurality of pixels of the acquired image into a plurality of groups based on the value of a variable in a predetermined color space of each pixel by a grouping processing routine. You may make it function. In addition, the CPU of the central control unit 10 functions as a variance value calculation unit that calculates, for each group, the variance value in the image of the pixels belonging to each group for each of the plurality of groups divided by the variance value calculation processing routine. You may make it let it. Further, the CPU of the central control unit 10 may function as an evaluation value calculating unit that calculates the evaluation value of each group based on the calculated dispersion value by the evaluation value calculation processing routine. In addition, the CPU of the central control unit 10 by the detection processing routine makes an area composed of pixels belonging to any one of a plurality of groups based on the evaluation value calculation result of each group as the attention area A1. You may make it function as a detection means to detect.
In addition, a program memory (not shown) that stores a program stores a program including an acquisition process routine, a grouping process routine, a center of gravity calculation process routine, a distance calculation process routine, an evaluation value calculation process routine, and a detection process routine. . Then, the CPU of the central control unit 10 is caused to function as a centroid calculating unit that calculates the centroid position of the pixel belonging to each group for each of the grouped groups by the centroid calculation processing routine. good. Further, the CPU of the central control unit 10 may function as a distance calculating unit that calculates the distance between the calculated center of gravity position and the center of the image by the distance calculation processing routine. Further, the CPU of the central control unit 10 may function as an evaluation value calculation unit that calculates the evaluation value of each group based on the calculated distance by the evaluation value calculation processing routine. Note that the functions of the acquisition processing routine, grouping processing routine, and detection processing routine are the same as described above, and a description thereof will be omitted.

100 imaging device 1 imaging unit 2 imaging control unit 5 image processing unit 5a image acquisition unit 5b smoothing unit 5c grouping unit 5d dispersion value calculation unit 5e centroid calculation unit 5f distance calculation unit 5g evaluation value calculation unit 5h attention area detection unit 5i Comparison determination unit 5j Output control unit 10 Central control unit

Claims (10)

An acquisition means for acquiring an image;
Grouping means for grouping a plurality of pixels of the image acquired by the acquisition means into a plurality of groups based on values of variables in a predetermined color space of each pixel;
For each of a plurality of groups grouped by the grouping means, a variance value calculating means for calculating a variance value in the image of pixels belonging to each group for each group;
Evaluation value calculating means for calculating an evaluation value of each group based on the dispersion value calculated by the dispersion value calculating means;
Based on a calculation result by the evaluation value calculation unit, a detection unit that detects a region constituted by pixels belonging to any one of the plurality of groups as a region of interest;
When a new attention area is detected by this detection means, a comparison is made with a predetermined number of past detection results, and it is determined whether or not to adopt the newly detected attention area according to the comparison result. A determination means;
An image processing apparatus comprising: An acquisition means for acquiring an image;
Grouping means for grouping a plurality of pixels of the image acquired by the acquisition means into a plurality of groups based on values of variables in a predetermined color space of each pixel;
For each of a plurality of groups grouped by the grouping means, a variance value calculating means for calculating a variance value in the image of pixels belonging to each group for each group;
Evaluation value calculating means for calculating an evaluation value of each group based on the dispersion value calculated by the dispersion value calculating means;
Based on a calculation result by the evaluation value calculation unit, a detection unit that detects a region constituted by pixels belonging to any one of the plurality of groups as a region of interest;
When the attention area detected by the detection means is composed of a plurality of closed areas, an exclusion means for excluding a closed area smaller than a predetermined size from the attention area;
Images it is processing device according to claim Rukoto equipped with. An acquisition means for acquiring an image;
Grouping means for grouping a plurality of pixels of the image acquired by the acquisition means into a plurality of groups based on values of variables in a predetermined color space of each pixel;
For each of the grouping means a plurality of groups that are by Ri grouped into a dispersion value calculating means for calculating a dispersion value in the image of the pixels belonging to each group for each group,
Based on the variance value calculated by the variance value calculating means, the evaluation value calculating means for calculating an evaluation value of each group,
Based on a calculation result by the evaluation value calculation unit, a detection unit that detects a region constituted by pixels belonging to any one of the plurality of groups as a region of interest;
Bei to give a,
Wherein each of the plurality of groups, the image processing apparatus according to the predetermined characteristic of Rukoto such overlap the group adjacent ranges of each component in the color space defining each group. An acquisition means for acquiring an image;
Grouping means for grouping a plurality of pixels of the image acquired by the acquisition means into a plurality of groups based on values of variables in a predetermined color space of each pixel;
For each of the plurality of groups grouped by the grouping means, a centroid calculating means for calculating the centroid position of the pixels belonging to each group for each group;
Distance calculating means for calculating the distance between the center of gravity calculated by the center of gravity calculating means and the center of the image;
Evaluation value calculation means for calculating an evaluation value of each group based on the distance calculated by the distance calculation means;
Based on a calculation result by the evaluation value calculation unit, a detection unit that detects a region constituted by pixels belonging to any one of the plurality of groups as a region of interest;
When a new attention area is detected by this detection means, a comparison is made with a predetermined number of past detection results, and it is determined whether or not to adopt the newly detected attention area according to the comparison result. A determination means;
Images it is processing device according to claim Rukoto equipped with. An acquisition means for acquiring an image;
Grouping means for grouping a plurality of pixels of the image acquired by the acquisition means into a plurality of groups based on values of variables in a predetermined color space of each pixel;
For each of the plurality of groups grouped by the grouping means, a centroid calculating means for calculating the centroid position of the pixels belonging to each group for each group;
Distance calculating means for calculating the distance between the center of gravity calculated by the center of gravity calculating means and the center of the image;
Evaluation value calculation means for calculating an evaluation value of each group based on the distance calculated by the distance calculation means;
Based on a calculation result by the evaluation value calculation unit, a detection unit that detects a region constituted by pixels belonging to any one of the plurality of groups as a region of interest;
When the attention area detected by the detection means is composed of a plurality of closed areas, an exclusion means for excluding a closed area smaller than a predetermined size from the attention area;
Images processing device comprising: a. An acquisition means for acquiring an image;
Grouping means for grouping a plurality of pixels of the image acquired by the acquisition means into a plurality of groups based on values of variables in a predetermined color space of each pixel;
For each of the plurality of groups grouped by the grouping means, a centroid calculating means for calculating the centroid position of the pixels belonging to each group for each group;
Distance calculating means for calculating the distance between the center of gravity calculated by the center of gravity calculating means and the center of the image;
Evaluation value calculation means for calculating an evaluation value of each group based on the distance calculated by the distance calculation means;
Based on a calculation result by the evaluation value calculation unit, a detection unit that detects a region constituted by pixels belonging to any one of the plurality of groups as a region of interest;
With
Wherein each of the plurality of groups, images processing apparatus characterized by comprising overlapping said predetermined group of adjacent ranges of each component in the color space defining each group. An acquisition step of acquiring an image;
A grouping step of grouping a plurality of pixels of the image acquired in this acquisition step into a plurality of groups based on values of variables in a predetermined color space of each pixel;
For each of a plurality of groups grouped in this grouping step, a variance value calculating step for calculating, for each group, a variance value in the image of pixels belonging to each group;
An evaluation value calculating step for calculating an evaluation value of each group based on the dispersion value calculated in the dispersion value calculating step;
Based on the definitive calculation result to the evaluation value calculation step, a detection step of detecting an area configured as a region of interest by any pixels that belong to one group of the plurality of groups,
When a region of interest is newly detected in this detection step, a comparison is made with a predetermined number of past detection results, and it is determined whether or not to adopt the newly detected region of interest according to the comparison result. A decision step to
The image processing method characterized by performing. An acquisition step of acquiring an image;
A grouping step of grouping a plurality of pixels of the image acquired in this acquisition step into a plurality of groups based on values of variables in a predetermined color space of each pixel;
For each of a plurality of groups grouped in this grouping step, a variance value calculating step for calculating, for each group, a variance value in the image of pixels belonging to each group;
Based on the variance value calculated by the variance value calculation step, an evaluation value calculation step of calculating an evaluation value of each group,
Based on the calculation result in the evaluation value calculation step, a detection step of detecting a region constituted by pixels belonging to any one of the plurality of groups as a region of interest;
An image processing method for executing
Each of the plurality of groups is formed by overlapping each component range in the predetermined color space defining each group with an adjacent group . Computer
Acquisition means for acquiring images,
Grouping means for grouping a plurality of pixels of the image acquired by the acquisition means into a plurality of groups based on the values of variables in a predetermined color space of each pixel;
For each of the grouping means a plurality of groups grouped by the centroid calculation means for calculating the center of gravity of the pixels belonging to each group for each group,
Distance calculating means for calculating the distance between the center of gravity calculated by the center of gravity calculating means and the center of the image ;
Based on the distance calculated by the distance calculating means, evaluation value calculating means for calculating an evaluation value of each group,
Detecting means for detecting, as a region of interest, an area constituted by pixels belonging to any one of the plurality of groups based on a calculation result by the evaluation value calculating means;
Exclusion means for excluding a closed area smaller than a predetermined size from the attention area when the attention area detected by the detection means is composed of a plurality of closed areas;
A program characterized by functioning as Computer
Acquisition means for acquiring images,
Grouping means for grouping a plurality of pixels of the image acquired by the acquisition means into a plurality of groups based on the values of variables in a predetermined color space of each pixel;
For each of a plurality of groups grouped by this grouping means, a centroid calculating means for calculating the centroid position of pixels belonging to each group for each group,
Distance calculating means for calculating the distance between the center of gravity calculated by the center of gravity calculating means and the center of the image;
Evaluation value calculation means for calculating an evaluation value of each group based on the distance calculated by the distance calculation means;
Detecting means for detecting, as a region of interest, an area constituted by pixels belonging to any one of the plurality of groups based on a calculation result by the evaluation value calculating means;
A program that functions as
Each of the plurality of groups is a program in which the range of each component in the predetermined color space defining each group overlaps with an adjacent group .

Images