JP4874167B2 - Image processing apparatus and imaging apparatus equipped with the same - Google Patents

Description

  The present invention relates to an image processing apparatus capable of detecting a specific object such as a face and an imaging apparatus equipped with the image processing apparatus.

  Digital video cameras have become widespread, and it has become easier for general users to shoot moving images than ever before. A general user often photographs a moving subject such as a child running at an athletic meet. At that time, there is an increasing demand for clear recording of subjects such as facial expressions of running children. On the other hand, a digital video camera equipped with a mode for imaging and recording at a very high frame rate, such as 300 frames per second, has been put into practical use.

Patent Document 1 discloses an imaging device that uses face detection technology.
JP 2006-233791 A

  As a matter of course, a moving image captured with high image quality presses the recording capacity. This is particularly noticeable in moving images captured at a high frame rate as described above. Therefore, a technique for efficiently encoding a moving image captured with high image quality is required.

  The present invention has been made in view of such circumstances, and an object of the present invention is to provide an image processing apparatus capable of reducing the amount of codes while satisfying the user's shooting intention, and an imaging apparatus equipped with the image processing apparatus.

  An image processing apparatus according to an aspect of the present invention includes an object detection unit that detects a predetermined object from a captured moving image, a control unit that adaptively changes a frame rate according to a detection result by the object detection unit, Is provided.

  It should be noted that any combination of the above-described constituent elements and a conversion of the expression of the present invention between a method, an apparatus, a system, a recording medium, a computer program, and the like are also effective as an aspect of the present invention.

  According to the present invention, the amount of codes can be reduced while satisfying the user's intention.

  The embodiment of the present invention suppresses the increase in the code amount of the entire moving image by recording the subject that the user is paying attention to with high image quality and recording the other areas with lower image quality than the subject. Technology. This technique is particularly effective in high-speed shooting mode and continuous shooting mode. The high-speed shooting mode is a mode for shooting at a very high frame rate such as 300 frames per second. A moving image captured in this mode becomes a smooth image even when slow playback is performed. The continuous shooting mode is a mode for continuously capturing still images. The resolution of each still image is set higher than a single image that composes a moving image shot in the moving image mode. As one of the continuous shooting modes, there is a spiral continuous shooting mode in which continuous shooting is performed while changing imaging conditions. This mode is used for searching for an optimum imaging condition.

  A problem common to the high-speed shooting mode and the continuous shooting mode is that the amount of code when recording a captured image increases. The embodiment of the present invention can effectively solve this problem. Of course, the embodiment of the present invention can also be applied to shooting in other modes, but the following description will focus on examples applied to the high-speed shooting mode. Further, although an embodiment of the present invention uses an object detection technique, an example in which a human face is detected as an object detection will be described. Note that the detection of an object is not limited to detection of a human face, but may be detection of a pet's face such as a dog or cat, or detection of an object such as a car, train or ship.

  FIG. 1 is a configuration diagram of an imaging apparatus 500 according to the first embodiment. The imaging apparatus 500 according to Embodiment 1 includes an imaging unit 10, an image processing unit 100, a recording unit 16, an operation unit 18, and a display unit 19. The imaging unit 10 includes an imaging element 12 and a signal processing unit 14. The image processing unit 100 includes a control unit 20, an encoding unit 22, and a face detection unit 24. The configuration of the image processing unit 100 can be realized in terms of hardware by a CPU, memory, or other LSI of an arbitrary computer, and is realized in terms of software by a program loaded in the memory. Describes functional blocks realized through collaboration. Therefore, those skilled in the art will understand that these functional blocks can be realized in various forms by hardware only, software only, or a combination thereof.

  The imaging device 12 is configured by a CCD (Charge Coupled Devices) sensor or a CMOS (Complementary Metal-Oxide Semiconductor) image sensor, converts a received signal into an electrical signal, and outputs the electrical signal to the signal processing unit 14.

  The signal processing unit 14 converts the RGB format analog signal output from the image sensor 12 into a YUV format digital signal. The signal processing unit 14 outputs the converted image signal to the image processing unit 100. More specifically, the data is output in parallel to the encoding unit 22 and the face detection unit 24 in units of frames.

  The face detection unit 24 detects the face of a person from the moving image captured and input by the imaging unit 10. The face detection unit 24 has a plurality of pattern data for detecting a person as dictionary data, searches a target frame for an area that matches or approximates the pattern data, and detects the face of the person by matching. .

  Here, the face detection may be performed by a known method, and is not particularly limited. For example, a face detection method based on an edge detection method, a boosting method, a hue extraction method, or a skin color extraction method can be used. The edge detection method extracts various edge features from the face area including the face, eyes, nose, mouth, face outline, etc. of the face image that has been normalized in advance for face size and gray value, and identifies whether it is a face A face discriminator is constructed by learning feature quantities that are effective for this purpose based on statistical methods.

  In order to detect a face from the input image, similar feature amounts are extracted while performing raster scanning from the end of the input image with the face size normalized during learning. From the feature amount, the discriminator determines whether the region is a face. As the feature amount, for example, a horizontal edge, a vertical edge, a right oblique edge, a left oblique edge, or the like can be used. If no face is detected, the input image is reduced at a certain rate, and the face is searched for the reduced image while performing raster scanning in the same manner as described above. By repeating such processing, a face of any size can be found in the image.

  Further, although the accuracy is lower than that of the edge detection method, the boosting method may be used when high-speed processing is desired. The boosting method detects a face from an image by comparing a shadow of a face such as an eye and nose with a shadow of a face detection pattern registered in advance without using an edge.

  The following methods may be used as other face detection methods. A method for extracting a face candidate area, dividing the face candidate area into small areas, comparing the feature amount of each area with a preset face area pattern, and extracting a face area from the accuracy, or a face candidate A method may be used in which a region is extracted and the face region is extracted by evaluating the accuracy from the degree of overlap of each candidate region. Furthermore, when face candidate areas are extracted and the density of each candidate area is a value corresponding to a predetermined threshold, the body candidate areas are extracted, and accuracy is determined using the density or saturation contrast of the face and body candidate areas. May be used to extract a face region.

  The face detection unit 24 notifies the control unit 20 of the result of the face detection process. In this embodiment, the presence / absence of a face may be notified for each frame. Further, instead of notifying every frame, it may be notified only when the presence or absence of a face changes. In this case, the processing amount can be reduced. The face detection process does not necessarily have to be performed for every frame, and may be performed every few frames. Also in this case, the processing amount can be reduced.

  The control unit 20 adaptively changes the frame rate according to the detection result by the face detection unit 24. Here, the frame rate may be any of the number of frames to be imaged per unit time, the number of frames to be encoded per unit time, and the number of frames to be recorded per unit time. In the case of the number of frames to be imaged per unit time, the control unit 20 instructs the imaging unit 10 to set the number of shutters to a predetermined value. In the case of the number of frames to be encoded or the number of frames to be recorded, the imaging unit 10 instructs the encoding unit 22 to set the ratio of the number of frames to be encoded or the ratio of the frames to be recorded.

  When the control unit 20 shifts from a state in which a face is detected in a moving image to a state in which a face is not detected as adaptive control, the control unit 20 switches to a frame rate lower than the frame rate in the state in which the face is detected. On the other hand, when the state is shifted from the state where no face is detected to the state where a face is detected, the frame rate is switched from the low frame rate to the state where the face is detected.

  Assuming the high-speed shooting mode, the frame rate when the face is detected is set to a high frame rate, such as 300 frames per second, because it is recorded with high image quality. The frame rate in a state where no face is detected is set to a lower frame rate than the state in which a face is detected, such as 60 frames per second and 120 frames per second, because there is little need to record so high.

  The encoding unit 22 compresses and encodes the image signal input from the signal processing unit 14 in accordance with a predetermined standard. For example, the standards of the MPEG series (MPEG-1, MPEG-2 and MPEG-4) standardized by the ISO (International Organization for Standardization) / IEC (International Electrotechnical Commission), and the international standards organizations related to telecommunications H.264 standardized by an ITU-T (International Telecommunication Union-Telecommunication Standardization Sector). H.26x series standards (H.261, H.262 and H.263), or H.264, which is a video compression coding standard standardized jointly by both standards organizations. H.264 / AVC (official recommendation names in both organizations are MPEG-4 Part 10: Advanced Video Coding and H.264 respectively).

  The encoding unit 22 records the encoded moving image data in the recording unit 16. In response to an instruction from the control unit 20, the encoding unit 22 changes the ratio of frames to be encoded or frames to be recorded.

  The recording unit 16 includes a memory card slot, an optical disc, or an HDD, and records a captured moving image on a recording medium. The operation unit 18 includes a shutter button and various setting buttons or keys, receives an instruction from the user, and transmits the instruction to the control unit 20. In the present embodiment, when an instruction is received from the user to operate in the high-speed shooting mode, the instruction is transmitted to the control unit 20. The display unit 19 displays an image being picked up and an image recorded in the recording unit 16, and displays various setting screens and various status information.

  FIG. 2 is a diagram illustrating a state in which the frame rate is adaptively switched depending on the presence or absence of face detection. An image 30, an image 32, and an image 34 show images that compose a moving image, and are shown in order of time passage. The image 30 shows a state where the person A is not present in the frame and the face of the person A is not detected. The image 32 shows a state in which the person A exists in the frame, and a degree of coincidence exceeding a predetermined set value is obtained between the area F including the face of the person A and one of the pattern data. The image 34 shows a state in which the person A does not exist in the frame and the face of the person A is not detected.

  The frame rate of the image 32 including the face of the person A is set higher than the frame rate of the images 30 and 34 not including the face of the person A. For example, before the face of the person A fits in the frame, the control unit 20 sets the frame rate to 60 frames per second or 120 frames per second, and when the face of the person A fits in the frame, the frame rate is set to 300 frames per second. When the person A's face is out of the frame, the frame rate is set again to 60 frames per second or 120 frames per second.

  FIG. 3 is a flowchart showing the operation of the image processing unit 100 according to the first embodiment. The face detection unit 24 searches the pattern data for detecting the face from the frames composing the input moving image and searches for a region having a matching degree exceeding a predetermined set value (S10). The control unit 20 determines the presence / absence of face detection (S12), and if detected (Y of S12), controls the encoding unit 22 to record at a high frame rate (S14), and cannot detect (N of S12) ), The encoder 22 is controlled to record at a low frame rate (S16).

  As described above, according to the first embodiment, the code amount can be reduced while satisfying the user's requirement for image quality by adaptively changing the frame rate according to the presence or absence of face detection. That is, recording is performed at a high frame rate when a face is detected, and recording is performed at a low frame rate when no face is detected. The face is an object that the user wants to record with high image quality. In the first embodiment, it is possible to record a scene with a face with high image quality and reduce the amount of codes in a scene without a face.

  This is particularly effective in high-speed shooting mode and continuous shooting mode where there is a strong demand for code amount reduction. For example, the high-speed shooting mode is suitable for sports shooting, and if you start shooting before the subject appears, you can start recording with high image quality as soon as it appears while reducing the amount of code before the subject appears. .

  FIG. 4 is a configuration diagram of an imaging apparatus 600 according to the second embodiment. The imaging apparatus 600 according to the second embodiment has the same basic part as the imaging apparatus 500 according to the first embodiment. Hereinafter, differences will be described.

  In addition to the components of the image processing unit 100 according to the first embodiment, the image processing unit 110 according to the second embodiment includes a detection mode / evaluation value conversion table 26 (hereinafter referred to as a first table) and an evaluation value. / Frame rate conversion table 28 (hereinafter referred to as the second table).

  The first table 26 associates and manages face detection modes and evaluation values. The first table 26 has different evaluation values depending on the shooting mode. That is, depending on the shooting mode, the same face detection mode has different evaluation values. The second table 28 manages the evaluation value and the frame rate in association with each other. The face detection unit 24 specifies not only the presence / absence of face detection but also the face detection mode. A specific example of the face detection mode will be described later. The control unit 20 refers to the first table 26 and the second table 28, specifies the frame rate according to the face detection mode, and instructs the encoding unit 22 to switch to the frame rate from the current frame rate. .

  FIG. 5 is a diagram illustrating an example of the first table 26. Here, the face detection mode is represented by four items of “number of face detections”, “probability of face”, “face orientation”, and “face size”. It should be noted that at least one of these may be included as an item representing the face detection mode.

  FIG. 5A is a diagram illustrating an example of the first table 26 in the potade mode (set to one person). “Number of detected faces” is an item for evaluating the number of detected faces in the same screen. Since the Po Trade mode (set to one person) is a mode for shooting one person, a state where no face is detected on the same screen or two or more faces are detected in the state intended by the user. The evaluation value is 0 points. The state in which one face is detected in the same screen is the intended state, and the evaluation value is a maximum of 5 points.

  “Faisiness of face” is an item for evaluating the degree of coincidence with pattern data. When the degree of coincidence is less than 0.5, the evaluation value is 0 point, when the degree of matching is 0.5 or more and 0.8 or less, the evaluation value is 3 points, and when it exceeds 0.8, the evaluation value is 5 points. Of course, the higher the degree of coincidence, the higher the score.

  The “face orientation” is an item for evaluating whether the face is facing the front or the side. In order to detect the face orientation, front face pattern data and side face pattern data may be prepared in advance and collated with each other. Occasionally, the front face may be determined when the number of detected eyes is two, and the side face may be determined when the number is one. The evaluation value is 5 points for the front face and 3 points for the side face. Of course, the front face has a higher score.

  “Face size” is an item for evaluating the relative size of the entire screen. The “face size” is obtained by calculating the ratio between the size of the entire screen and the size of the face detected in the screen. When the area occupied by the face with respect to the entire screen is less than 2%, the evaluation value is 1 point. When the area is 2% or more and 10% or less, the evaluation value is 2 points. When the area exceeds 10%, the evaluation value is 3 points. Naturally, the larger the score, the higher the score. The control unit 20 converts the face detection mode in the target image into an evaluation value by adding all the scores of each item.

  FIG. 5B is a diagram illustrating an example of the first table 26 in the potade mode {set to n (n is an integer of 2 or more)}. With regard to the “number of face detections”, the po trade mode (set to n persons) is a mode for photographing n persons, so that other than n in the same screen, {0 to (n−1), n + 1 <}. The state in which the face is detected is not the state intended by the user, and the evaluation value is 0 for all the detected faces. A state in which n faces are detected in the same screen is an intended state, and the maximum evaluation value is 5 for all the detected faces.

  The “probability of face”, “face orientation”, and “face size” are the same as those in the portray mode (set to one person) shown in FIG. In this mode, for each detected face, the control unit 20 uses a score obtained from the item “number of detected faces”, a score obtained from the item “probability of face”, and an item “face orientation”. The obtained score and the score obtained from the item “face size” are added. Furthermore, all the calculated total scores for each face are added together, and the score is divided by the number of detections. This score is taken as the final evaluation value. Accordingly, the face detection mode can be converted into the evaluation value while removing the influence due to the difference in the number of detections from the evaluation value.

  FIG. 5C shows an example of the first table 26 in the group photo mode. As the score in the item “number of detected faces”, 10 points are given as evaluation values for each detected face. The evaluation of “probability of the face” is the same as in the case of the potade mode (set to n persons) shown in FIG. For the “face orientation”, the score of the face is multiplied by 1.0 when the face is a front face, and 0.5 is multiplied by the score of the face when the face is a side face. “Face size” is multiplied by 0.5 when the area of the face occupying the entire screen is less than 1%, and multiplied by 0.7 when the face score is between 1% and 2%. If it exceeds 2%, multiply the score on that face by 1.0. In the group photo, each face is originally small, but the evaluation is lowered by setting the weight to less than 1 only when it is less than 2%.

  In this mode, for each detected face, the control unit 20 adds the score obtained from the item “number of detected faces” and the score obtained from the item “probability of face”, By multiplying the magnification obtained from the item “face orientation” and the magnification obtained from the item “face size”, a score for each detected face is calculated. Further, all the calculated scores for each face are added together, and the score is divided by the number of detections. This score is taken as the final evaluation value. Accordingly, the face detection mode can be converted into the evaluation value while removing the influence due to the difference in the number of detections from the evaluation value. In this mode, the evaluation value is set higher as the number of face detections increases. This is because in the group photo, the larger the number of face detections, the closer to the intended state.

  Note that the contents of the items for evaluating the face detection mode shown in FIG. 5, the scores for the evaluation results of the items, and the weighting of the scores for each mode are examples, and the designer can arbitrarily set them. For example, the presence or absence of blinking may be included in the item for evaluating the face detection mode.

  FIG. 6 is a diagram illustrating an example of the second table 28. The second table 28 is a table for converting the evaluation value obtained in the first table 26 into a frame rate. Here, the frame rate in the high-speed shooting mode and the continuous shooting mode with respect to the evaluation value is shown.

  In the high-speed shooting mode, the frame rate is 30 [fps] when the evaluation value is in the range of 0 to 5, the frame rate is 60 [fps] when the evaluation value is in the range of 5 to 10, and the frame rate is within the range of the evaluation value is 10 to 15. Is set to 120 [fps], and the frame rate is set to 300 [fps] in an evaluation value range of 15 to 20.

  In the continuous shooting mode, the frame rate is 1 [fps] when the evaluation value is 0 to 5, the frame rate is 3 [fps] when the evaluation value is 5 to 10, and the frame rate is 10 to 15 when the evaluation value is 10 to 15. Is 5 [fps], and the frame rate is set to 10 [fps] in the range of 15 to 20 evaluation values.

  The higher the evaluation value, the closer to the state that the user desires to record with high image quality, such as a state in which the person to be the subject is facing up and appearing up. Therefore, the higher the evaluation value, the higher the frame rate is set and the higher the image quality is recorded.

  FIG. 7 is a diagram illustrating how the frame rate is switched in accordance with the evaluation value. An image 40, an image 42, and an image 44 show images that compose a moving image, and are shown in order of time passage. The image 40 shows a state where the person A is not present in the frame and the face of the person A is not detected. The image 42 shows a state in which the person A exists in the frame and the person A faces the side. The image 44 shows a state in which the person A is present in the frame and the person A is facing the front.

  When the evaluation value is calculated with reference to the first table 26 as shown in FIG. 5, the evaluation value of the image 40 is the lowest, the evaluation value of the image 44 is the highest, and the evaluation value of the image 42 is between them. . For example, in the high-speed shooting mode, the control unit 20 sets the frame rate to 30 [fps] when switching to the image 40, sets to 120 [fps] when switching to the image 42, and switches to the image 44. Set to 300 [fps].

  FIG. 8 is a flowchart showing the operation of the image processing unit 110 according to the second embodiment. The face detection unit 24 detects a face and its detection mode from within each frame composing the input moving image, and the control unit 20 refers to the first table 26 to determine the face detection mode as an evaluation value. Conversion is performed (S20). Next, the evaluation value is converted into a frame rate with reference to the second table 28 and set in the encoding unit 22 (S22). While there is no shooting end instruction (N in S24), the control unit 20 transitions to Step S20 and repeats the same processing. In the first table 26, the frame rate is maintained as long as the evaluation value remains within the range of each class. When the evaluation value changes beyond the range of each class, the frame rate is changed.

  As described above, according to the second embodiment, the balance between the image quality and the code amount can be optimized as compared with the first embodiment by changing the frame rate according to the evaluation value of the face detection mode. For example, record a scene that is large and facing the front with the highest image quality, record a scene that is small and facing the side with a lower image quality, and record a scene with no face in the lowest image quality. It is possible. By such processing, it is possible to reduce the code amount while satisfying the user's request for image quality.

  Also, by setting an evaluation value for each shooting mode, the code amount can be reduced while satisfying the user's shooting intention. For example, in the po trade mode for single shooting, the situation in which a plurality of persons are captured in the image is not the situation intended by the user. In this case, if the evaluation value is set to be low, the amount of unnecessary code increases. Can be suppressed. On the other hand, in the group photo mode for multi-person photographing, a situation where only a small number of faces are shown in the image, such as a face hidden behind a person in the front row, is not a situation intended by the user. In the group photo mode, if the evaluation value is set so as to increase as more faces appear in the image, an unnecessary increase in code amount can be suppressed while satisfying the user's intention.

  FIG. 9 is a configuration diagram of an imaging apparatus 700 according to the third embodiment. The imaging apparatus 700 according to the third embodiment has the same configuration as that of the imaging apparatus 600 according to the second embodiment, and differences will be described below.

  The image processing unit 120 according to the third embodiment includes an evaluation value / bit rate conversion table 29 instead of the evaluation value / frame rate conversion table 28 of the image processing unit 120 according to the second embodiment.

  The first table 26 associates and manages face detection modes and evaluation values. In Embodiment 3, it is not necessary to calculate the evaluation value of the entire image, and the evaluation value may be calculated for each area where the face is detected. Therefore, it is not necessary to add the evaluation values for each face and divide by the number of detections. The evaluation value / bit rate conversion table 29 manages the evaluation value and the bit rate of the region of interest (ROI) in association with each other. Here, the bit rate is a bit rate at the time of converting the pixel value of the region of interest into an orthogonal transform coefficient such as a discrete cosine transform coefficient or a wavelet transform coefficient. Similar to the second table 28, the bit rate is designed to be higher as the evaluation value is higher.

  The encoding unit 22 sets a region including a face as a region of interest. For example, it may be set to the same size as the window frame for the face search, or may be set larger by a predetermined number of pixels vertically and horizontally. When a plurality of faces are detected, an attention area may be set for each face, or attention areas may be set for some of the detected faces. The encoding unit 22 converts the pixel value of the region of interest at a bit rate instructed by the control unit 20 when converting the pixel value of the region of interest into an orthogonal transform coefficient. When the high bit rate is specified, the attention area can be recorded with high image quality, and when the low bit rate is specified, the code amount of the attention area can be reduced.

  The control unit 20 refers to the first table 26 and the evaluation value / bit rate conversion table 29, specifies the bit rate according to the face detection mode, and uses the bit rate to determine the pixel value of the region of interest including the face. Instructs the encoding unit 22 to convert to a coefficient. In addition, when a plurality of faces are detected in the same frame, an attention area is set for each face, and the pixel value of each attention area is converted into a coefficient at a bit rate corresponding to each face detection mode. An instruction is given to the encoding unit 22.

  As described above, according to the third embodiment, the image quality and code amount of the object to be noticed are changed by changing the bit rate when converting the pixel value into the coefficient according to the evaluation value of the face detection mode. The balance can be optimized. For example, it is possible to record the face facing the front with the highest image quality, record the face facing the side with a lower image quality, and record the non-attention area with the lower image quality. By such processing, it is possible to reduce the code amount while satisfying the user's request for image quality.

  The present invention has been described based on some embodiments. It is understood by those skilled in the art that these embodiments are exemplifications, and that various modifications can be made to combinations of the respective constituent elements and processing processes, and such modifications are also within the scope of the present invention. By the way.

  In the embodiment described above, a human face is assumed as a face detection target, but an animal face such as a dog or a cat may be used. If the pattern data for dogs and the pattern data for cats are registered, it can be carried out according to the same principle as the case of a human face. Note that pattern data of pets such as dogs and cats may be generated by a user imaging a pet and extracting feature points or the like from the image within the imaging apparatus. In this case, it is possible to easily generate highly accurate pattern data for one's pet.

1 is a configuration diagram of an imaging apparatus according to Embodiment 1. FIG. It is a figure which shows a mode that a frame rate is switched adaptively with the presence or absence of face detection. 3 is a flowchart illustrating an operation of the image processing unit according to the first embodiment. 3 is a configuration diagram of an imaging apparatus according to Embodiment 2. FIG. FIG. 5A is a diagram illustrating an example of the first table 26 in the potade mode (set to one person). FIG. 5B is a diagram illustrating an example of the first table 26 in the potade mode {set to n (n is an integer of 2 or more)}. FIG. 5C shows an example of the first table 26 in the group photo mode. It is a figure which shows an example of a 2nd table. It is a figure which shows a mode that a frame rate is switched according to an evaluation value. 10 is a flowchart illustrating an operation of an image processing unit according to the second embodiment. 6 is a configuration diagram of an imaging device according to Embodiment 3. FIG.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 10 Image pick-up part, 12 Image pick-up element, 14 Signal processing part, 16 Recording part, 18 Operation part, 19 Display part, 20 Control part, 22 Encoding part, 24 Face detection part, 26 1st table, 28 2nd table, 100 An image processing unit, 500 an imaging device.

Claims (4)

An object detection unit for detecting a predetermined object from the captured moving image;
A control unit that adaptively changes a frame rate of the moving image according to a detection result by the object detection unit;
Equipped with a,
The control unit changes the frame rate of the moving image according to the size of the detected object, and shifts from the state in which the object is detected in the moving image to the state in which the object is not detected. An image processing apparatus that switches to a frame rate lower than a frame rate in a state where the object is detected . A first table associating a detection mode of the object with an evaluation value;
A second table associating the evaluation value with a frame rate;
The image processing apparatus according to claim 1 , wherein the control unit switches to a frame rate according to a detection mode of the object with reference to the first table and the second table . The image processing apparatus according to claim 2 , wherein the first table has a different evaluation value depending on a shooting mode . An image sensor for capturing a moving image;
The image processing apparatus according to claim 1, which processes a moving image captured by the image sensor.
An imaging apparatus comprising:

Images