JP2013016924A - Image evaluation device - Google Patents

Abstract

An image evaluation apparatus capable of determining whether a main subject is out of frame or not is an object of image evaluation.
A storage unit that stores image data, an extraction unit that extracts a main subject region including a main subject from an image based on the image data, and whether or not the main subject is out of frame in the image data. A setting unit 10 that sets a frame-out area for determining whether or not the image is to be set along at least one edge of the image, and the main subject is a frame when at least part of the main subject area overlaps the frame-out area. The determination part 10 which determines with having been out is provided.
[Selection] Figure 1

Description

  The present invention relates to an image evaluation apparatus.

  2. Description of the Related Art Conventionally, there has been known an image photographing apparatus that performs image evaluation such as exposure, contrast, and blur on a photographed image and deletes image data of an image whose evaluation value is equal to or less than a predetermined threshold from a recording medium (for example, Patent Document 1).

JP 2006-050494 A

  However, in the above-described image capturing apparatus, whether or not the main subject is out of frame is not subject to image evaluation.

  An object of the present invention is to provide an image evaluation apparatus that can determine whether or not a main subject is out of frame.

An image evaluation apparatus according to the present invention includes a storage unit that stores image data, an extraction unit that extracts a main subject region including a main subject from an image based on the image data, and the main subject in the image data is framed out. A setting unit for setting a frame-out region for determining whether or not the image is present along at least one edge of the image; and when the main subject is overlapped with the frame-out region, And a determination unit that determines that the frame is out.
An image evaluation apparatus according to the present invention includes a storage unit that stores a plurality of image data acquired in time series, an extraction unit that extracts a main subject region including a main subject from an image based on the image data, and the image data A setting unit for setting a frame-out region for determining whether or not the main subject is out of frame along at least one edge of the image; and at least a part of the main subject region is the frame-out region A determination unit that determines that the main subject is out of frame in a case where the main subject is out of frame, and an evaluation unit that reduces an evaluation value of the image data determined to be out of frame by the determination unit. To do.

  According to the image evaluation apparatus of the present invention, whether or not the main subject is out of frame can be set as the object of image evaluation.

It is a block diagram which shows the system configuration | structure of the electronic camera which concerns on 1st Embodiment. It is a flowchart which shows the process of the electronic camera which concerns on 1st Embodiment. It is a figure which shows the image data acquired in time series in the electronic camera which concerns on 1st Embodiment. It is a flowchart which shows the evaluation process of the image data in the electronic camera which concerns on 1st Embodiment. It is a figure which shows the image based on the image data acquired by the electronic camera which concerns on 1st Embodiment. It is a figure which shows the image based on the image data acquired by the electronic camera which concerns on 1st Embodiment. It is a figure which shows the image based on the image data acquired by the electronic camera which concerns on 1st Embodiment. It is a figure which shows the image based on the image data acquired by the electronic camera which concerns on 1st Embodiment. It is a flowchart which shows the evaluation process of the image data in the electronic camera which concerns on 2nd Embodiment. It is a figure which shows the image based on the image data acquired by the electronic camera which concerns on 2nd Embodiment. It is a figure which shows the image based on the image data acquired by the electronic camera which concerns on 2nd Embodiment. It is a figure which shows the image based on the image data acquired by the electronic camera which concerns on 2nd Embodiment. It is a figure which shows the image based on the image data acquired by the electronic camera which concerns on 2nd Embodiment. It is a flowchart which shows the evaluation process of the image data in the electronic camera which concerns on 3rd Embodiment. It is a figure which shows the image based on the image data acquired by the electronic camera which concerns on 3rd Embodiment. It is a flowchart which shows the evaluation process of the image data in the electronic camera which concerns on 4th Embodiment. It is a figure which shows the image based on the image data acquired by the electronic camera which concerns on 4th Embodiment. It is a figure which shows the image based on the image data acquired by the electronic camera which concerns on 4th Embodiment. It is a figure which shows the image based on the image data acquired by the electronic camera which concerns on 4th Embodiment. It is a flowchart which shows the evaluation process of the image data in the electronic camera which concerns on 5th Embodiment. It is a figure for demonstrating the frame-out area | region used with the electronic camera which concerns on 5th Embodiment. It is a figure which shows the case where it is determined with the electronic camera which concerns on 5th Embodiment that the main to-be-photographed object has gone out of frame. It is a figure for demonstrating the frame-out area | region used with the electronic camera which concerns on 6th Embodiment. It is a figure which shows the case where it is determined with the electronic camera which concerns on 6th Embodiment that the main to-be-photographed object has gone out of frame. It is a figure for demonstrating the frame-out area | region used by the electronic camera which concerns on embodiment. It is a figure which shows the case where it is determined with the electronic camera which concerns on embodiment that the main to-be-photographed object has gone out of frame. It is a figure which shows the case where it is determined with the electronic camera which concerns on embodiment that the main to-be-photographed object has gone out of frame. It is a figure for demonstrating the frame-out area | region used by the electronic camera which concerns on embodiment. It is a figure for demonstrating the frame-out area | region used by the electronic camera which concerns on embodiment. It is a flowchart which shows the process of the electronic camera which concerns on 7th Embodiment. It is a flowchart which shows the process of the electronic camera which concerns on 8th Embodiment.

  Hereinafter, an electronic camera as an image evaluation apparatus according to the first embodiment will be described with reference to the drawings. FIG. 1 is a block diagram showing a system configuration of the electronic camera 2 according to the first embodiment. As shown in FIG. 1, the electronic camera 2 includes a control unit 10 that is configured by a microprocessor or the like and controls each part of the electronic camera 2 in an integrated manner. An imaging device 12, LCD display unit 14, buffer memory 16, information storage unit 18, data storage unit 20, program memory 22, memory card 24, and operation unit 30 are connected to the control unit 10.

  The image sensor 12 is configured by a CMOS image sensor or the like, images light from a subject via a photographing lens (not shown), and outputs an image signal to the control unit 10. The LCD display unit 14 is disposed on the back side of the electronic camera 2 and displays a through image or the like based on the imaging signal from the imaging device 12. The buffer memory 16 temporarily records a plurality of frames of image data based on the imaging signal output from the imaging device 12. The information storage unit 18 stores information necessary for determining whether or not the main subject is out of frame in the image data (hereinafter referred to as frame out determination). The data storage unit 20 stores a predetermined file (hereinafter referred to as an evaluation file) that stores an evaluation value for each file name of image data. The program memory 22 stores a program for the control unit 10 to comprehensively control each unit of the electronic camera 2 and to perform each process such as an imaging process, a reproduction process, and an image selection process. The memory card 24 is a portable storage medium that is detachably mounted in a card slot (not shown) provided in the electronic camera 2, and is, for example, a CF card, an SD card, a smart media, or the like.

  The operation unit 30 includes a power switch (not shown) for turning on / off the power of the electronic camera 2, a half-push switch SW1 that is turned on by depressing the shutter button to a half-pressed position, A push switch SW2 is provided.

  Next, processing in the electronic camera 2 according to the first embodiment will be described with reference to FIG. First, the operator operates a button (not shown) of the electronic camera 2 to select optimal image data from among a plurality of image data continuously acquired in time series as the operation mode of the electronic camera 2. A selection mode is set (step S1). Next, the control unit 10 starts imaging by the image sensor 12 and displays a through image on the LCD display unit 14 based on the image signal read from the image sensor 12 at a predetermined frame rate (for example, 30 fps).

  Next, when the shutter button is pressed down to the half-pressed position and the half-press switch SW1 is turned on (S1) (step S2), the control unit 10 starts the exposure control operation (AE) and the focus adjustment operation (AF). (Step S3). That is, an exposure control operation (AE) is performed based on the image pickup signal output from the image pickup device 12, and the shutter speed and the aperture value are determined so as to obtain an appropriate exposure. For example, when detecting the brightness (luminance) information of a subject by the average photometry method, a value obtained by averaging the signal values constituting the through image is used as the subject luminance information, and based on the luminance information of the subject. Determine the optimum exposure value and the shutter speed and aperture value. Further, the control unit 10 obtains a defocus amount by focus detection processing by pupil division using the focus detection signal read from the image sensor 12, and performs focusing of a photographing lens (not shown). Here, it is assumed that a focus detection pixel based on a pupil division phase difference method is incorporated in the CMOS image sensor constituting the imaging device 12.

  When the exposure control operation (AE) and the focus adjustment operation (AF) are completed, the control unit 10 starts pre-capture of image data (step S4). That is, the control unit 10 continuously acquires image data of a plurality of frames as pre-captured image data based on the imaging signal output from the image sensor 12 and records the acquired image data of the plurality of frames in the buffer memory 16. Here, the number of image data of a plurality of frames to be recorded in the buffer memory 16 is determined to be, for example, 10 frames as the pre-capture setting number as shown in FIG. Therefore, when the number of image data recorded in the buffer memory 16 reaches 10 frames, the control unit 10 deletes the old image data in order. Thereby, the capacity of the buffer memory 16 used for pre-capture can be saved.

  Next, the control unit 10 determines whether or not the release button is pressed down to the fully pressed position and the fully pressed switch SW2 is turned on (S2) (step S5). When the full press switch SW2 is not turned on (step S5: No), the pre-capture of the image data is continued (step S4).

  When the full push switch SW2 is turned on (S2) (step S5: Yes), the control unit 10 starts post-capturing of image data (step S6). That is, the control unit 10 includes a plurality of frames of image data including image data (hereinafter referred to as reference image data) captured when the full-press switch SW2 is turned on based on an imaging signal output from the imaging element 12. Are continuously acquired as post-capture image data, and the acquired plural frames of image data are recorded in the buffer memory 16. Here, the number of image data of a plurality of frames to be recorded in the buffer memory 16 is determined as 10 frames including the image data captured at the time when the switch SW2 is turned on (S2), for example, as the post-capture setting number. (See FIG. 3).

  When the post-capture is completed, the control unit 10 includes image data recorded in the buffer memory 16 (that is, 10-frame image data including 10-frame image data acquired by pre-capture and reference image data acquired by post-capture). Evaluation is performed on 20 frames of image data (step S7).

  The image data evaluation process (step S7) in the electronic camera 2 of the first embodiment will be described below with reference to FIG. First, the control unit 10 reads the reference image data 51 (see FIG. 3) from the buffer memory 16 (step S11), and detects a main subject (for example, a car) included in the image based on the reference image data 51 (step S12). . Here, the detection of the main subject is performed by a generally known subject recognition technique.

  Next, the control unit 10 extracts information on the characteristics of the main subject (hereinafter referred to as main subject information) from the image based on the reference image data 51 (step S13). For example, when a car is detected as a main subject, the size and shape of the car are extracted as main subject information. Next, the control unit 10 stores the main subject information in the information storage unit 18. The main subject information stored in the information storage unit 18 is used when a main subject is detected in an image based on other image data.

  Next, the control unit 10 determines a region including the main subject (hereinafter referred to as a main subject region) in the image based on the reference image data 51 (step S14). For example, as shown in FIG. 5, a rectangular area including a car that is a main subject is determined as a main subject area 70. Next, the control unit 10 stores information on the main subject region such as the position and size of the main subject region 70 in the information storage unit 18. The main subject area 70 is used for determining the moving direction of the main subject.

  Next, the control unit 10 stores, as an evaluation value of the reference image data 51, an evaluation value when the main subject is not out of frame (hereinafter referred to as a reference evaluation value) in the evaluation file of the data storage unit 20 (step). S15). Here, the reference evaluation value is stored in the evaluation file in association with the file name of the reference image data 51. As shown in FIG. 5, in the image based on the reference image data 51, the main subject region 70 is not in contact with the edge of the image, and the main subject is not out of frame.

  Next, the control unit 10 reads the image data 52 (see FIG. 3) acquired next to the reference image data 51 from the buffer memory 16 (step S16), and reads main subject information from the information storage unit 18. Then, the main subject corresponding to the main subject detected in step S12 is detected from the image based on the image data 52 using the main subject information (step S17). Next, the control unit 10 determines a main subject area in the image based on the image data 52 (step S18).

  Next, the control unit 10 determines whether or not the image data read in step S16 is image data acquired next to the reference image data 51 (step S19). Here, since the image data 52 is image data acquired next to the reference image data 51 (step S19: Yes), the control unit 10 determines the moving direction of the main subject (step S20). That is, the control unit 10 reads information on the main subject area of the reference image data 51 from the information storage unit 18, and the main subject area determined in step S 18 is located on either side of the main subject area of the image based on the reference image data 51. The moving direction of the main subject is determined by determining whether it is located. For example, as shown in FIG. 6, when the main subject area 72 of the image data 52 is located on the lower left side of the main subject area 70 of the reference image data 51, the lower left direction is determined as the moving direction of the main subject. . Next, the control unit 10 stores information related to the moving direction of the main subject in the information storage unit 18.

  On the other hand, if the image data read in step S16 is not the image data acquired next to the reference image data 51 (step S19: No), the moving direction of the main subject has already been determined, so the control unit 10 newly Does not determine the direction of movement.

  Next, the control unit 10 performs frame out determination on the image data 52 (step S21). Here, the frame-out determination is performed by determining whether or not at least a part of the edge of the main subject region 72 on the moving direction side is in contact with the edge of the image. For example, when the moving direction of the main subject is the lower left side direction, as shown in FIG. 7, when the left edge and the lower edge of the main subject area 72 are in contact with the edge of the image, It is determined that the subject is out of frame. As shown in FIG. 8, only the left edge of the main subject area 72 is in contact with the edge of the image, or only the lower edge of the main subject area 72 is in contact with the edge of the image. Even if it is present (not shown), it is determined that the main subject is out of frame. On the other hand, when the left edge and the lower edge of the main subject area 72 are not in contact with the edge of the image, it is determined that the main subject is not out of frame.

  When the main subject is not out of frame (step S21: No), the control unit 10 does not change the evaluation value, and stores the reference evaluation value in the evaluation file in association with the file name of the image data 52 ( In step S22), it is determined whether or not all the image data recorded in the buffer memory 16 have been evaluated (step S24). Note that when the image data 52 acquired by the post-capture is evaluated, the image data acquired by the pre-capture is not evaluated (step S24: No), and the process returns to the step S16. And the process after step S16 is performed about the image data 53 (refer FIG. 3) acquired after the image data 52. FIG. When the main subject of the image data 53 is not out of the frame (step S21: No), the control unit 10 does not change the evaluation value and associates the reference evaluation value with the file name of the image data 53 in the evaluation file. After the storage (step S22), the process returns to step S16 again (step S24: No), and the same process is repeated for the image data acquired by post-capture.

  On the other hand, when the main subject is out of frame (step S21: Yes), the control unit 10 stores an evaluation value lower than the reference evaluation value in the evaluation file in association with the file name of the image data 52 (step S23). ).

  In the frame-out determination according to the present embodiment, the control unit 10 determines the moving direction of the main subject, and is acquired after the image data in which the moving portion side edge of the main subject region is in contact with the edge of the image. For the image data, the frame out determination is not performed, and an evaluation value lower than the reference evaluation value is stored in the evaluation file.

  Further, the control unit 10 determines the moving direction of the main subject, and the image data acquired before the image data in which the edge opposite to the moving direction of the main subject region is in contact with the edge of the image is obtained. The evaluation value lower than the reference evaluation value is stored in the evaluation file without performing the frame-out determination.

  Next, the control unit 10 determines whether or not all of the image data recorded in the buffer memory 16 has been evaluated (step S24). When image data acquired by pre-capture has not been evaluated (step S24: No), the process returns to step S16, and for example, for image data 50 (see FIG. 3) acquired by pre-capture, step S16 and subsequent steps. Perform the process. Note that the frame-out determination for the image data acquired by the pre-capture is performed by at least a part of the edge opposite to the moving direction side of the main subject area determined in step S20 (for example, the moving direction of the main subject is the lower left side) In some cases, the determination is made by determining whether at least one of the right side and the upper side of the main subject area is in contact with the edge of the image. If the main subject of the image data 50 is not out of frame (step S21: No), the control unit 10 does not change the evaluation value, and evaluates the reference evaluation value in association with the file name of the image data 50. After storing in the file (step S22), the process returns to step S16 again (step S24: No), and the same process is repeated for the pre-captured image data acquired before the image data 50.

  When all of the image data recorded in the buffer memory 16 has been evaluated (step S24: Yes), the control unit 10 reads the evaluation file from the data storage unit 20, and evaluates the image data (FIG. 2, step S7). To complete. Then, image data in which the reference evaluation value is stored in the evaluation file (hereinafter referred to as image data of the reference evaluation value) is selected as a recorded image and recorded on the memory card 24 (FIG. 2, step S8).

  In the above-described embodiment, the image data captured when SW2 is turned on is used as the reference image data. However, the first image of pre-capture may be used as the reference image data. Alternatively, the position of the main subject in the screen may be detected from the recorded image, and an image in which the position is located in the center of the screen is used as reference image data, and the main subject may be detected based on the reference image data. . With this configuration, it is possible to set a main subject that is not out of frame with certainty as reference image data, and the frame out determination system is improved.

  According to the electronic camera 2 according to the first embodiment, whether or not the main subject is out of frame can be set as the object of image evaluation for a plurality of pieces of image data acquired in time series. In addition, the moving direction of the main subject is determined, and the image data acquired after (before) the image data in which the edge on the moving direction side of the main subject region is in contact with the edge of the image is determined to be out of frame. The evaluation value lower than the reference evaluation value is stored in the evaluation file. For this reason, for example, when a main subject with a clear movement direction is imaged while the electronic camera 2 is fixed to a tripod or the like, image data in which the main subject is not out of frame can be selected efficiently.

  Further, in the above-described embodiment, the frames before and after the time when the main subject is out of the frame and the main subject is switched out of the frame or the main subject is switched out of the frame. If the image data acquired after (before) the image is determined based on the detected moving direction of the main subject and the frame-out position of the main subject, the frame-out determination of the main subject is performed. Can be performed more efficiently.

  Next, processing of the electronic camera according to the second embodiment will be described. In the first embodiment, the electronic camera according to the second embodiment does not determine the moving direction of the main subject but performs frame-out determination for all the image data recorded in the buffer memory 16 in the first embodiment. It is a thing. Therefore, in the second embodiment, a part relating to the image data evaluation process, which is a different part from the first embodiment, will be described in detail, and the description of the overlapping part will be omitted as appropriate.

  FIG. 9 is a flowchart showing image data evaluation processing in the electronic camera according to the second embodiment. First, the control unit 10 reads the reference image data 51 (see FIG. 3) from the buffer memory 16 (step S31), and detects the main subject included in the image based on the reference image data 51 (step S32).

  Next, the control unit 10 extracts main subject information from the image based on the reference image data 51 (step S33) and stores it in the information storage unit 18. Next, the control unit 10 determines the main subject area in the image based on the reference image data 51 (step S34), determines the number of contact sides of the main subject area that touches the edge of the image (step S35), Information related to the number of sides is stored in the information storage unit 18. For example, as shown in FIG. 10, when the main subject region 80 in the image based on the reference image data 51 is in contact with the edge of the image, the number of contact sides is determined as “1”. Also, as shown in FIG. 11, when the main subject region 80 is not in contact with the edge of the image, the number of contact sides is determined as “0”. Next, the control unit 10 stores the reference evaluation value in the evaluation file of the data storage unit 20 in association with the file name of the reference image data 51 (step S36).

  Next, the control unit 10 reads the image data 52 (see FIG. 3) acquired next to the reference image data 51 from the buffer memory 16 (step S37), and reads main subject information from the information storage unit 18. Next, the control unit 10 detects the main subject corresponding to the main subject detected in step S32 from the image based on the image data 52 using the main subject information (step S38). Next, the control unit 10 determines a main subject region in the image based on the image data 52 (step S39), and determines the number of contact sides of the main subject region that are in contact with the edge of the image (step S40).

  Next, the control unit 10 reads out information on the number of contact sides of the reference image data 51 from the information storage unit 18 and compares the number of contact sides in the reference image data 51 with the number of contact sides in the image data 52. Thus, frame out determination is performed on the image data 52 (step S41). For example, as shown in FIG. 12, when the number of tangent sides in the image data 52 is larger than the number of tangent sides in the reference image data 51, it is determined that the main subject is out of frame in the image data 52. Further, as shown in FIG. 13A, when the number of tangent sides in the image data 52 is smaller than the number of tangent sides in the reference image data 51, or as shown in FIG. When the number of tangent sides is the same as the number of tangent sides in the reference image data 51, it is determined that the main subject is not out of frame in the image data 52.

  If the main subject is not out of frame (step S41: No), the control unit 10 does not change the evaluation value, and stores the reference evaluation value in the evaluation file in association with the file name of the image data 52 (step). S42). On the other hand, when the main subject is out of frame (step S41: Yes), the control unit 10 stores an evaluation value lower than the reference evaluation value in the evaluation file in association with the file name of the image data 52 (step S43). ).

  Next, the control unit 10 determines whether or not all of the image data recorded in the buffer memory 16 has been evaluated (step S44). When all the image data recorded in the buffer memory 16 has not been evaluated (step S44: No), the process returns to the process of step S37, and the step is performed for the image data 53 (see FIG. 3) acquired after the image data 52. The process of S37-44 is performed. Thereafter, the control unit 10 repeats the processes of steps S37 to S44 for the image data acquired by post capture. When the evaluation is completed for all the image data acquired by the post-capture, the control unit 10 performs the processes of steps S37 to S44 for the image data 50 (see FIG. 3) acquired by the pre-capture, for example. Thereafter, the processes in steps S37 to S44 are repeated for the pre-captured image data acquired before the image data 50.

  When all of the image data recorded in the buffer memory 16 has been evaluated (step S24: Yes), the control unit 10 reads the evaluation file from the data storage unit 20, and evaluates the image data (FIG. 2, step S7). The image data of the reference evaluation value is selected as a recorded image and recorded on the memory card 24 (FIG. 2, step S8).

  According to the electronic camera of the second embodiment, since all of the image data recorded in the buffer memory 16 is evaluated, for example, a moving subject is panned and the main subject tends to be seen only from the image. Even when the moving direction cannot be specified, image data in which the main subject is not out of frame can be selected accurately.

  Next, processing of the electronic camera according to the third embodiment will be described. In the electronic camera according to the third embodiment, in the second embodiment, instead of determining the number of tangent sides of the main subject area, at least a part of the main subject area is behind the other subject. It is determined whether it is hidden or not. The frame-out determination is performed based on the determination result. Therefore, in the third embodiment, processing different from that of the second embodiment will be described in detail, and description of overlapping portions will be omitted as appropriate.

  FIG. 14 is a flowchart showing image data evaluation processing in the electronic camera according to the third embodiment. First, the control unit 10 reads the reference image data 51 (see FIG. 3) from the buffer memory 16 (step S51), and detects the main subject included in the image based on the reference image data 51 (step S52).

  Next, the control unit 10 extracts main subject information from the image based on the reference image data (step S53) and stores it in the information storage unit 18. Next, the control unit 10 stores the reference evaluation value in the evaluation file of the data storage unit 20 in association with the file name of the reference image data 51 (step S54).

  Next, the control unit 10 reads the image data 52 (see FIG. 3) acquired next to the reference image data 51 from the buffer memory 16 (step S55), and reads main subject information from the information storage unit 18. Next, the control unit 10 detects the main subject corresponding to the main subject detected in step S52 from the image based on the image data 52 using the main subject information (step S56). Next, the control unit 10 determines a main subject area in the image based on the image data 52 (step S57).

  Next, the control unit 10 detects another subject different from the main subject in the image based on the image data 52 (step S58), and performs frame-out determination on the image data 52 based on the detection result (step S59). For example, when another subject different from the main subject is detected in the image based on the image data 52, the control unit 10 determines whether at least a part of the main subject region is hidden behind the other subject. That is, as shown in FIG. 15, when a part of the main subject area 84 in the image based on the image data 52 is hidden behind another subject 86, the control unit 10 frames the main subject out of the image data 52. It is determined that On the other hand, when the main subject area is not hidden behind other subjects or when no other subject is detected in the image based on the image data 52, the control unit 10 frames the main subject out of the image data 52. Judge that it is not.

  The processing in steps S60 to 62 is the same as the processing in steps S42 to S44 in the second embodiment, and a description thereof will be omitted.

  According to the electronic camera of the third embodiment, another subject different from the main subject is detected, and the main subject is a frame for image data of an image in which at least a part of the main subject area is hidden behind the other subject. Since it is determined that the main subject is out, it is possible to accurately select image data of a good image in which the main subject is clearly displayed.

  Next, processing of the electronic camera according to the fourth embodiment will be described. In the electronic camera according to the fourth embodiment, in the second embodiment, instead of determining the number of tangent sides of the main subject area, at least a part of one main subject area is the other main subject area. It is determined whether or not the subject area overlaps. Then, frame out determination is performed based on the determination result. Therefore, in the fourth embodiment, processing different from that of the second embodiment will be described in detail, and description of overlapping portions will be omitted as appropriate.

  FIG. 16 is a flowchart showing image data evaluation processing in the electronic camera according to the fourth embodiment. First, the control unit 10 reads the reference image data 51 (see FIG. 3) from the buffer memory 16 (step S71), and detects a plurality of main subjects included in the image based on the reference image data 51 (step S72). For example, as shown in FIG. 17, when a plurality of persons are included in the image, the face portions of the persons are detected as the main subject 94 and the main subject 96, respectively.

  Next, the control unit 10 extracts the main subject information of the main subject 94 and the main subject information of the main subject 96 from the image based on the reference image data 51 (step S73), and stores them in the information storage unit 18. Next, the control unit 10 stores the reference evaluation value in the evaluation file of the data storage unit 20 in association with the file name of the reference image data 51 (step S74).

  Next, the control unit 10 reads the image data 52 (see FIG. 3) acquired next to the reference image data 51 from the buffer memory 16 (step S75), and the main subject information of the main subject 94 from the information storage unit 18. , And main subject information of the main subject 96 is read out. Next, the control unit 10 detects a plurality of main subjects from the image based on the image data 52 using the main subject information read from the information storage unit 18 (step S76). That is, as shown in FIG. 18, a main subject 97 corresponding to the main subject 94 and a main subject 98 corresponding to the main subject 96 are detected. Next, the control unit 10 determines a main subject region 99 including the main subject 97 and a main subject region 100 (see FIG. 18) including the main subject 98 in the image based on the image data 52 (step S77).

  Next, the control unit 10 performs frame-out determination by determining whether or not at least a part of one main subject area overlaps with another main subject area in the image based on the image data 52 (step S78). For example, as shown in FIG. 19, when a part of the main subject region 99 overlaps the main subject region 100, it is determined that the main subject is out of frame. On the other hand, as shown in FIG. 18, when the main subject area 99 and the main subject area 100 do not overlap, it is determined that the main subject is not out of frame.

  The processing in steps S79 to 81 is the same as the processing in steps S42 to S44 in the second embodiment, and thus the description thereof is omitted.

  According to the electronic camera of the fourth embodiment, it is possible to accurately select image data of a good image even when there are a plurality of main subjects.

  Next, image data evaluation processing according to the fifth embodiment will be described. In the electronic camera according to the fifth embodiment, in the first to fourth embodiments, an area for performing frame-out determination (hereinafter referred to as a frame-out area) is set in an image. Accordingly, the part relating to the setting of the frame-out area, which is a part different from the first to fourth implementations, will be described in detail, and the description of the overlapping part will be omitted as appropriate.

  FIG. 20 is a flowchart showing image data evaluation processing in the electronic camera according to the fifth embodiment. First, the control unit 10 reads the reference image data 51 (see FIG. 3) from the buffer memory 16 (step S95), and detects a main subject (for example, a human face) included in the image based on the reference image data 51 (step S95). S96). Next, main subject information is extracted from the image based on the reference image data 51 (step S97) and stored in the information storage unit 18. Here, the main subject information is used when a main subject is detected in an image based on other image data.

  Next, the control unit 10 determines a main subject area in the image based on the reference image data 51 (step S98). For example, as shown in FIG. 21, a square area including a human face 60 as a main subject is determined as the main subject area 62.

  Next, the control unit 10 determines the width of the frame-out area based on the detected size of the main subject (step S99). For example, the control unit 10 measures the area of the person's face 60 (see FIG. 21) displayed in the image based on the reference image data 51, and the larger the area of the person's face 60, the wider the frame-out region, The smaller the area of the human face 60 is, the smaller the width of the frame-out region is determined.

  Next, in the image based on the reference image data 51, the control unit 10 sets the frame-out area having the width determined in step S99 (step S100). For example, when the width of the frame-out area determined in step S99 is the width A, as shown in FIG. 21, a U-shaped area having a width A along both the edge and the lower edge of the image. Is set as the frame-out area 64. Note that the position and shape (excluding the width) of the frame-out region 64 are set in advance by the operator.

  Next, the control unit 10 performs frame out determination in the reference image data 51 by determining whether or not at least a part of the main subject region 62 overlaps the frame out region 64 (step S101). For example, as shown in FIG. 22, when a part of the main subject area 62 overlaps the frame-out area 64, the control unit 10 determines that the main subject is out of frame. On the other hand, as shown in FIG. 21, when the main subject area 62 does not overlap the frame-out area 64, the control unit 10 determines that the main subject is not out of frame.

  In addition, about the process of step S102-S104, since it is the same as that of the process of step S22-24 in 1st Embodiment, description is abbreviate | omitted.

  According to the electronic camera 2 according to the fifth embodiment, whether or not the main subject is out of frame can be set as the object of image evaluation for a plurality of pieces of image data acquired in time series. Also, in order to increase the width of the frame-out area as the area of the main subject is larger, for example, when a human face is displayed larger in the image, if the face position is far from the center of the image, It is easy to determine that the subject is out of frame. On the other hand, since the width of the frame-out region is narrowed as the area of the main subject is small, for example, image data of an image in which a human face is displayed small is difficult to determine that the main subject is out of frame. Thereby, it is possible to accurately select image data having a well-balanced composition.

  Next, processing of the electronic camera according to the sixth embodiment will be described. In the electronic camera according to the sixth embodiment, in the fifth embodiment, the position of the main subject is further detected, and the width of the frame-out area is determined based on the position and size of the main subject. It is a thing. Then, when at least one of the main subjects overlaps the frame-out area, it is determined that the main subject is out of frame. Therefore, in the sixth embodiment, a part related to the image data evaluation process, which is a part different from the fifth embodiment, will be described in detail with reference to FIG. 20, and description of overlapping parts will be omitted as appropriate. To do.

  First, the control unit 10 reads the reference image data 51 (see FIG. 3) from the buffer memory 16 (step S95), and detects a plurality of main subjects included in the image based on the reference image data 51 (step S96). For example, as shown in FIG. 23, when a plurality of persons are included in the image, the face portions of the persons are detected as a main subject 72, a main subject 74, and a main subject 76, respectively. Next, main subject information is extracted from the image based on the reference image data 51 (step S97) and stored in the information storage unit 18.

  Next, the control unit 10 determines a main subject area corresponding to each main subject in the image based on the reference image data 51 (step S98). That is, as shown in FIG. 23, a main subject region 80 corresponding to the main subject 72, a main subject region 82 corresponding to the main subject 74, and a main subject region 84 corresponding to the main subject 76 are determined.

  Next, the control unit 10 determines the width of the frame-out area based on the position and size of the main subject (step S99). That is, the control unit 10 first detects the position of the main subject in the image. Next, the main subject closest to each edge of the image is determined. For example, in FIG. 23, the main subject closest to the left edge of the image is determined to be the main subject 72. Similarly, the main subject closest to the lower edge of the image is determined to be the main subject 74, and the main subject closest to the right edge of the image is determined to be the main subject 76.

  Next, the control unit 10 measures the area of the main subject, and determines the width of the frame-out region based on the area of the main subject closest to the edge on the side where the frame-out region is set. For example, when a rectangular frame-out region 86 (see FIG. 23) is set along the left edge of the image, the width B of the frame-out region 86 is based on the area of the main subject 72 closest to the frame-out region 86. To decide. Similarly, the width C of the frame-out region 88 is determined based on the area of the main subject 74 closest to the lower edge of the image, and based on the area of the main subject 76 closest to the right edge of the image. The width D of the frame-out area 90 is determined. Note that the width of the frame-out region is determined so as to increase as the area of the main subject increases and decrease as it decreases. Therefore, as shown in FIG. 23, when the area of the main subject becomes smaller in the order of the main subject 72, the main subject 76, and the main subject 74, the width of the frame-out area is the width B of the frame-out area 86, and the frame-out area. The width becomes narrower in the order of the width D of the area 90 and the width C of the frame-out area 88.

  Next, in the image based on the reference image data 51, the control unit 10 sets the frame-out area having the width determined in step S99 (step S100). That is, as shown in FIG. 23, a frame-out area 86 having a width B along the left edge of the image, a frame-out area 88 having a width C along the lower edge of the image, and the right edge of the image. A frame-out area 90 having a width D is set along the part.

  Next, the control unit 10 determines whether or not at least one of the main subject region 80, the main subject region 82, and the main subject region 84 overlaps the frame-out region. Out determination is performed (step S101). For example, as shown in FIG. 24, when a part of the main subject region 80 overlaps the frame-out region 86, the control unit 10 determines that the main subject is out of frame. On the other hand, as shown in FIG. 23, when no main subject area overlaps the frame-out area, the control unit 10 determines that the main subject is not out of frame.

  According to the electronic camera of the sixth embodiment, image data with a well-balanced composition can be accurately selected even when there are a plurality of main subjects.

  In the fifth embodiment, the orientation of the main subject may be determined, and the width of the frame-out area may be determined based on the orientation of the main subject. For example, the width of the frame-out region in the direction in which the main subject is directed may be widened, and the width of the frame-out region in the opposite direction in which the main subject is not directed may be narrowed. As a result, as shown in FIG. 25, when the face 94 of the person facing the right side is positioned at the left end of the image, the width X of the frame-out area 96 on the left side of the image is determined to be narrow. It becomes difficult to be judged that it is out. Further, since the width Y of the frame-out area 97 on the right side of the image is determined widely (see FIG. 25), if the face 94 of the person facing the right side is located at the right end of the image (not shown), It is easy to determine that the main subject is out of frame. For this reason, it is possible to accurately select image data having a well-balanced composition.

  In the fifth and sixth embodiments, it is further determined whether or not at least a part of the main subject overlaps the frame-out area, and when at least a part of the main subject overlaps the frame-out area. Alternatively, it may be determined that the main subject is out of frame. First, the control unit 10 determines whether or not the main subject area overlaps the frame-out area. If a part of the main subject area 100 overlaps the frame-out area 102 as shown in FIG. It is determined that the subject is out of frame. On the other hand, when the main subject region 100 does not overlap the frame-out region 102, the control unit 10 determines whether at least a part of the main subject 98 overlaps the frame-out region. As shown in FIG. 27, when a part of the main subject 98 overlaps the frame-out area, it is determined that the main subject is out of frame. Thereby, even when the size of the main subject area is smaller than the size of the entire main subject, image data with a well-balanced composition can be selected accurately.

  In the fifth and sixth embodiments, the width of the frame-out area may be determined based on the number of main subjects. First, the control unit 10 determines the number of main subjects detected from the image. When the number of main subjects is singular, the control unit 10 determines the width of the frame-out region (hereinafter referred to as a reference width) when the number of main subjects is singular as shown in FIG. Determine as F. On the other hand, when there are a plurality of main subjects, the control unit 10 determines, for example, a value obtained by dividing the reference width value by the number of main subjects as the width of the frame-out region. That is, as shown in FIG. 29, when there are three main subjects, a width that is 1/3 of the reference width (that is, the width F) is determined as the width F ′ of the frame-out area 108. As a result, the width of the frame-out area is determined so that the larger the number of main subjects, the narrower the frame-out area, so that even when there are a plurality of main subjects, image data with a well-balanced composition can be accurately obtained. Can be selected.

  Next, the image data evaluation process (FIG. 2, step S7) according to the seventh embodiment will be described. The electronic camera according to the seventh embodiment is configured such that the number of selected image data is a predetermined number as a result of the frame-out determination in the first to sixth embodiments. . Therefore, in the seventh embodiment, processes different from those in the first to sixth embodiments will be described in detail, and description of overlapping parts will be omitted as appropriate.

  FIG. 30 is a flowchart illustrating image data evaluation processing in the electronic camera according to the seventh embodiment. Here, FIG. 30 shows an example of processing of the electronic camera 2 according to the first embodiment, but the present embodiment can also be applied to the electronic camera according to the second to sixth embodiments.

  When the processing of steps S111 to 123 is completed and all the image data recorded in the buffer memory 16 is evaluated (step S124: Yes), the control unit 10 reads the evaluation file from the data storage unit 20. Then, it is determined whether or not the image data of the reference evaluation value is a predetermined number (for example, 5 frames) or less (step S125).

  When the image data of the reference evaluation value is equal to or less than the predetermined number (step S125: Yes), the control unit 10 completes the image data evaluation process (FIG. 2, step S7) and records the image data of the reference evaluation value. It selects as an image and records on the memory card 24 (FIG. 2, step S8).

  On the other hand, when the image data of the reference evaluation value is not equal to or smaller than the predetermined number (step S125: No), the control unit 10 performs the evaluation again on the image data of the reference evaluation value (step S126). For example, the sharpness of the image based on the image data is evaluated, and a new evaluation value is determined based on the sharpness. Then, the control unit 10 stores the new evaluation value in the evaluation file for each file name of the image data of the reference evaluation value, and completes the image data evaluation process (FIG. 2, step S7). Next, the control unit 10 selects a predetermined number (for example, 5 frames) of image data of reference evaluation values in order from the highest new evaluation value as a recorded image and records it on the memory card 24 (FIG. 2, step S8).

  According to the electronic camera 2 according to the seventh embodiment, for a plurality of image data acquired in time series, whether or not the main subject is out of frame can be set as an evaluation target of the image data. The image data can be evaluated and selected in consideration of other evaluation items. For this reason, it is possible to accurately select image data of a good image based on a plurality of evaluation items including evaluation based on frame-out determination. In addition, since image data is selected in consideration of evaluation items other than the evaluation based on the frame-out determination, the number of image data can be selected to a desired number.

  In the seventh embodiment, the case where the sharpness evaluation is performed as an evaluation other than the evaluation based on the frame-out determination is described as an example. However, the present invention is not limited to this, and an image such as a blur evaluation, an in-focus evaluation, etc. The evaluation may be performed based on any evaluation item or combination of evaluation items as long as the evaluation can evaluate the data. Further, these evaluation items or combinations of evaluation items may be configured to be automatically set for each scene mode preset by the operator in the electronic camera, or the type of main subject recognized by the electronic camera, etc. It may be set based on the characteristics of the image. Furthermore, it may be configured such that the operator can select an evaluation item.

  In this way, by setting an optimal evaluation item according to the image to be captured, image data is evaluated based on a plurality of evaluation items including evaluation based on frame-out determination, and an optimal recorded image is obtained for each scene. You can choose.

  Next, the image data evaluation process (FIG. 2, step S7) according to the eighth embodiment will be described. The electronic camera according to the eighth embodiment performs evaluation based on sharpness determination, blur determination, and in-focus determination in parallel with evaluation based on frame-out determination in the seventh embodiment. The image data is selected by comprehensive evaluation based on a plurality of evaluation items. Therefore, in the eighth embodiment, processing different from that in the seventh embodiment will be described in detail, and description of overlapping portions will be omitted as appropriate.

  FIG. 31 is a flowchart showing image data evaluation processing in the electronic camera according to the eighth embodiment. Here, FIG. 31 shows an example of processing of the electronic camera 2 according to the first embodiment, but the present embodiment can also be applied to the electronic camera according to the second to sixth embodiments.

  First, the control unit 10 performs evaluation based on sharpness determination, blur determination, and in-focus determination for image data (see FIG. 3) acquired in time series in parallel with evaluation based on frame-out determination. .

  Next, when all of the image data recorded in the buffer memory 16 are evaluated (step S144: Yes), the control unit 10 reads the evaluation file from the data storage unit 20, and performs evaluation based on frame-out determination, sharpness The evaluation value of the image data is comprehensively determined based on the four evaluation items of evaluation, blur evaluation, and focus degree evaluation (step S145).

  Next, the control unit 10 stores the determined evaluation value in the evaluation file for each file name of the image data, and completes the image data evaluation process (FIG. 2, step 7). Next, the control unit 10 selects the image data in the descending order of evaluation value (for example, 5 frames) as a recorded image and records it on the memory card 24 (FIG. 2, step S8).

  According to the electronic camera 2 according to the eighth embodiment, whether or not the main subject is out of frame can be used as the evaluation target of the image data for a plurality of image data acquired in time series. The image data can be evaluated and selected in consideration of other evaluation items. For this reason, the image according to the objective can be selected exactly based on a plurality of evaluation items including the evaluation based on the frame-out determination. In addition, since image data is selected in consideration of evaluation items other than the evaluation based on the frame-out determination, the number of image data can be selected to a desired number.

  In the eighth embodiment, a case where sharpness evaluation, blur evaluation, and focus degree evaluation are performed as evaluations other than the evaluation based on the frame-out determination is described as an example. The evaluation may be performed based on any evaluation item or combination of evaluation items as long as the evaluation can be evaluated. Further, these evaluation items or combinations of evaluation items may be configured to be automatically set for each scene mode preset by the operator in the electronic camera, or the type of main subject recognized by the electronic camera, etc. It may be configured to be set based on the characteristics of the image. Further, the operator may select an evaluation item or a combination of evaluation items.

  As described above, by setting the optimum evaluation item according to the image to be photographed, the image data can be evaluated based on the plurality of evaluation items, and the optimum recorded image can be selected for each scene.

  In the above-described embodiment, the case where the main subject region is a rectangular region has been described as an example. However, if the main subject region can be used to determine the frame-out determination or the moving direction of the main subject, the main subject region The shape of the region does not matter.

  In the above-described embodiment, the case where the image evaluation is performed in the electronic camera has been described as an example. However, the image data may be evaluated in another external device (for example, a personal computer). For example, the control unit 10 records the image data on the memory card 24 after the pre-capture and post-capture are completed for the image data of a plurality of frames recorded in the buffer memory 16 (FIG. 2, step S6). The operator removes the memory card 24 from the electronic camera and attaches it to the personal computer. Then, the control unit of the personal computer may read the image data from the attached memory card 24, and may perform evaluation and selection of the image data using the method described in the above embodiment.

DESCRIPTION OF SYMBOLS 2 ... Electronic camera, 10 ... Control part, 12 ... Imaging device, 14 ... LCD display part, 16 ... Buffer memory, 18 ... Information storage part, 20 ... Data storage part, 22 ... Program memory, 24 ... Memory card, 30 ... Operation part

Claims (7)

A storage unit for storing image data;
An extraction unit that extracts a main subject region including a main subject from an image based on the image data;
A setting unit that sets a frame-out region for determining whether or not the main subject is out of frame in the image data, along at least one edge of the image;
An image evaluation apparatus comprising: a determination unit that determines that the main subject is out of frame when at least a part of the main subject region overlaps the frame out region.   The image evaluation apparatus according to claim 1, further comprising a determination unit that determines a width of the frame-out area based on at least one of a size, a number, and an orientation of the main subject.   3. The image evaluation apparatus according to claim 2, wherein the determination unit increases the width of the frame-out area as the main subject is larger, and narrows the width of the frame-out area as the main subject is smaller.   3. The image according to claim 2, wherein the determination unit narrows the width of the frame-out region as the number of the main subjects increases, and increases the width of the frame-out region as the number of the main subjects decreases. Evaluation device.   The width of the frame-out area in the direction in which the main subject faces is widened, and the width of the frame-out area in the direction in which the main subject is not oriented is narrowed. Image evaluation device.   The image evaluation apparatus according to claim 1, wherein the main subject is a human face. A storage unit for storing a plurality of image data acquired in time series;
An extraction unit that extracts a main subject region including a main subject from an image based on the image data;
A setting unit that sets a frame-out region for determining whether or not the main subject is out of frame in the image data, along at least one edge of the image;
A determination unit that determines that the main subject is out of frame when at least a part of the main subject region overlaps the frame out region;
An image evaluation apparatus comprising: an evaluation unit that reduces an evaluation value of the image data determined to be out of frame by the determination unit.

Images