JP2010283609A - Imaging apparatus, face detection processing method, program and recording medium - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an imaging apparatus which prevents deviation of a face display frame that occurs in face detection processing and accelerates the detection by reducing wasteful processings. <P>SOLUTION: The imaging apparatus includes: an imaging means for inputting image data of a photographed object; a field angle change detecting means for detecting that a field angle of the image data is changed; and a face detecting means for detecting a person's face from the photographed image by using the image data obtained by the imaging means, wherein the face detecting means once stops operation processing, acquires a new image and performs operation processing for face detection, when the field angle change detecting means detects a change in the field angle during the operation processing for face detection. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to an imaging apparatus, a face detection processing method, a program, and a recording medium, and more particularly to a technique that is preferably applied to face detection processing in an imaging apparatus such as a digital camera or a digital video camera.

  In recent years, in an imaging device such as a digital camera, a function of detecting a person's face from a captured image and controlling the detected face area to be in focus, exposure, or the like, and the detected face are registered in advance. A function to authenticate whether or not the face is present is realized. However, when trying to realize these functions, the processing takes time, and if the camera itself is moved during detection or the subject moves, the face detection frame differs from the actual face position. It may be displayed at a place or there may be a time lag in following the face.

  For example, in Patent Document 1, a face position motion amount is predicted from a frame image in which a face is detected based on a change in camera pan / tilt amount obtained by an angular velocity sensor, and a face detection frame is defined based on the detected position. By correcting the display position, the face detection frame is prevented from being shifted from the face position of the current monitoring image.

  However, in the method according to Patent Document 1, if the face position motion amount prediction from the camera pan / tilt amount change deviates from the actual motion amount, the face detection frame display deviates from the actual face position. It will end up.

  SUMMARY OF THE INVENTION An object of the present invention is to prevent a face display frame from being shifted in a face detection process in an image pickup apparatus, and to reduce useless processing to speed up detection.

  An imaging apparatus according to one aspect of the present invention is obtained by an imaging unit for inputting image data of a photographed subject, an angle-of-view change detecting unit for detecting that the angle of view of the image data has changed, and an imaging unit. Face detection means for detecting a person's face from the captured image using the obtained image data. The face detection means detects a change in the view angle by the view angle change detection means during the face detection calculation process. If so, the calculation process is once stopped, a new image is acquired, and the face detection calculation process is performed.

  The face detection processing method according to one aspect of the present invention is obtained by an imaging step of acquiring image data of a photographed subject, an angle-of-view change detecting step of detecting a change in the angle of view of the image data, and an imaging step. A face detection step for detecting a human face from the photographed image using the captured image data. The face detection step detects a change in the angle of view during the face detection calculation process. If so, the calculation process is once stopped, a new image is acquired, and the face detection calculation process is performed.

  A program according to one aspect of the present invention includes a field angle change detection function for detecting that a field angle of image data has changed in a computer of an imaging apparatus having an imaging unit for inputting image data of a photographed subject. And a face detection function for detecting a person's face from the captured image using the image data obtained by the imaging means, and the face detection function is performed by a field angle change detection function during the face detection calculation process. When a change in the angle of view is detected, the calculation process is stopped once, a new image is acquired, and the face detection calculation process is performed.

  A recording medium according to one aspect of the present invention is a computer-readable recording medium on which the above program is recorded.

  According to the present invention, in the imaging apparatus, it is possible to prevent the face display frame from being shifted in the face detection process and reduce the useless process to speed up the detection.

1 is an external view of an imaging apparatus according to an embodiment of the present invention. 1 is a functional block diagram illustrating a schematic configuration of an imaging apparatus according to an embodiment of the present invention. It is the flowchart which showed the flow of the face detection process which concerns on embodiment of this invention. It is a figure for demonstrating the face detection process which concerns on embodiment of this invention. It is a figure for demonstrating the face detection process which concerns on embodiment of this invention. It is the flowchart which showed the flow of the face detection process which concerns on embodiment of this invention. It is the flowchart which showed the flow of the face detection process which concerns on embodiment of this invention.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings.

  FIG. 1 is an external view of a digital still camera according to the present embodiment, and FIG. 2 is a system schematic diagram inside the digital still camera. First, an outline of the operation of the digital still camera will be described.

  When the shooting / playback switching mode dial 3 in FIG. 1 is set to the shooting mode and the power button 1 is pressed, the camera is activated in the recording mode. The mode dial 3 and the power button 1 are included in the operation unit 60 in FIG. 2. When the CPU 43 detects that the power button 1 is turned on while the mode dial 3 is in the shooting mode, the CPU 43 The driver 20 is controlled to move the lens barrel unit including the zoom / AF lens 11, the diaphragm and the color filter 12, and the mechanical shutter 13 to a photographing position. Furthermore, power is turned on to the image sensor 30, the LCD display, and the like to start the operation.

  When the power of each part is turned on, the operation in the finder mode is started. In the finder mode, light incident on the image sensor (CCD, CMOS) 30 through the lens is converted into electricity and sent to the A / D converter 32. The A / D converter 32 converts the input analog signal into 12-bit digital data. Each signal converted into a digital signal is taken into the sensor I / F unit 41 in the digital signal processing IC 40, converted into a YUV signal in a format that can be displayed by the YUV conversion unit 44, and then framed by the memory controller 42. It is written in the SDRAM 50 as a memory. The YUV signal is read by the memory controller 42 and sent to the TV or LCD monitor via the display output control unit 46 for display. This process is performed at 1/30 second intervals, and a finder mode display updated every 1/30 seconds is obtained.

  The imaging sensor 30 is an 8 million pixel sensor. During still image shooting, all pixels can be output every 1/30 seconds. In the finder mode, output is performed by reducing the number of pixels to 640 × 480 pixels every 1/30 seconds by combining peripheral pixels with addition and thinning.

  The sensor I / F 41 of the digital signal processing IC 40 detects an AF evaluation value indicating the degree of focus on the screen, an AE evaluation value that detects the subject brightness, and an AWB that detects the subject color from the digital RGB signal captured in the block. An evaluation value is calculated. These data are read out as feature data to the CPU 43 and used for AE (Automatic Exposure), AF (Auto Focus), and AWB (Auto White Balance) processes.

  The AF evaluation value is created by, for example, an output integrated value of a high frequency component extraction filter or an integrated value of a luminance difference between adjacent pixels. When in the in-focus state, the edge portion of the subject is clear, so the high frequency component is the highest. By utilizing this, at the time of the focus detection operation by AF, the AF evaluation value at each focus lens position is acquired, and AF is executed with the point where the maximum is obtained as the focus position.

  The AE evaluation value and the AWB evaluation value are created from the respective RGB integration values. For example, the screen is equally divided into 256 areas (16 horizontal divisions and 16 vertical divisions), and the RGB integration of each area is calculated. The CPU 43 reads the RGB integration values, and in AE, the luminance of each area is calculated, and an appropriate exposure time is determined from the luminance distribution. In AWB, an AWB control value that matches the color of the light source is determined from the RGB distribution. The AE and AWB processes are continuously performed during the finder mode.

  When the release shutter button 2 in FIG. 1 is operated, an AF operation that is a focus position detection and a still image recording process are performed. When the release shutter button 2 is pressed, when a still image shooting start signal is fetched into the CPU 43 from the operation unit 20 in FIG. 2, the CPU 43 drives the lens via the motor driver 20 in synchronization with the frame rate, thereby increasing the hill-climbing AF. Execute. When the AF target range is the entire region from infinity to close, the focus lens moves to each focus position from close to infinity, or from infinity to close, and each frame (= each) created by the digital signal processing IC 40 The CPU 43 reads the AF evaluation value at the (focus position). The focus lens is moved to the in-focus position with the point at which the AF evaluation value at each focus position is maximized as the in-focus position.

  It is converted into a digital RGB signal output from the image sensor 30 after the AF is completed, and stored in the SDRAM 50 via the digital signal processing IC 40. The digital RGB signal is read again into the digital signal processing IC 40, converted into YUV data by the YUV conversion unit 44, and written back to the SDRAM 50.

  At the time of capturing a still image, YUV converted image data is sent to the compression / decompression circuit 45 in the digital signal processing IC 40. The YUV data sent to the compression / decompression circuit 45 is compressed and written back to the SDRAM 50. The compressed data of the SDRAM 50 is read through the digital signal processing IC 40 and stored in the data storage memory.

The acceleration sensor 90 is mounted on the PCB and outputs data of two axes X and Y and temperature T. The tilt of the camera roll angle, pitch angle, etc. is calculated from the data and displayed on an LCD monitor or the like. The roll angle θ with respect to the horizontal of the acceleration sensor is expressed by the following equation.
θ [deg] = (180 / π) * arctan (Y0−G0 / X0−G0)
G0 is an output at the time of zero gravity.
When the camera is in the lower position in FIG. 1, the roll angle θ of the camera is assumed to be 0 degree, and when the LCD monitor is tilted clockwise, the tilt is positive, and when the camera is tilted counterclockwise, the tilt is negative. Suppose there is.

  Next, a face detection method will be described. Many methods are known for detecting a human image from a subject image. In this embodiment, for example, the method shown in the following reference is used. The following reference describes a method for detecting a face by calculating a Haar-Like feature value based on a classifier by creating a cascade combination of a number of weak classifiers in cascade by AdaBoost learning. ing.

(References)
P. Viola and M.M. Jones, “Rapid Object Detection using a Boosted Cascade of Simple Features”, Proc. IEEE International Conference on Computer Vision and Pattern Rec., Vol. 1, pp. 511-518, 2001

  Hereinafter, in this embodiment, a weak classifier in the above method is described as feature data, and a classifier obtained by linearly combining weak classifiers in a cascade form is described as a feature data group. As for the face tracking method based on the face detection result of the previous cycle, tracking is realized by performing the face detection process limited to a small range around the face detected by the face detection of the previous cycle. Shall.

[First Embodiment]
Next, face detection processing, which is an operation that characterizes the present invention, will be described in three embodiments. FIG. 3 is a face detection process corresponding to the first embodiment. The face detection process is a flow of detecting a person's face from the image by performing the face detection process on the image captured by the imaging optical system. It is.

  When the camera operation mode is a mode for performing face detection processing, the face detection module copies one image from the monitoring image at a certain timing (step S101). The CPU 43 performs face detection processing on the copied image and determines whether or not there is a human face in the image. In the face detection process, the similarity between the human face feature data stored in the memory in advance and the symmetric rectangle in the image is calculated. In this embodiment, the feature data in the memory is divided into a plurality of layers. It is assumed that similarity of feature data is calculated for each hierarchy.

  Here, for example, in the feature data, in two or three small rectangles in the target rectangle as shown in FIG. 5, when the weighted sum of the average luminance values of the small rectangles is equal to or greater than a threshold value, It is data that it is. In this embodiment, the weighted sum of the average luminance values of each small rectangle is called a feature evaluation value. The feature data in the shallow hierarchy is a relatively rough judgment as to whether or not the target rectangle is a face. For example, as shown in FIG. 5A, a large small rectangle in the target rectangle is retained. The number of feature data is also small. The deeper the hierarchy, the smaller the small rectangles as shown in FIGS. 5 (b) and 5 (c), so that the fine features of the face are judged, and the number of feature data in the hierarchy also increases. The similarity in each hierarchy is calculated by obtaining a feature evaluation value in each feature data in the hierarchy. The more the feature data whose feature evaluation value exceeds the threshold value, the higher the similarity. If the feature evaluation value of the feature data j in the hierarchy i is Evj, and the similarity is simj when Evj is equal to or greater than the threshold Thj, the similarity Sim of the hierarchy i is calculated by the following equation.

  First, the first target rectangle is determined from the image (step S102). Here, it is the upper left edge of the image. In the present embodiment, as shown in FIG. 4, scanning is performed while gradually shifting the target rectangular area in order from the upper left end to the lower right end of the image.

  First, the hierarchy L = 1 is set (step S103), and the similarity between the feature data of the first hierarchy and the target rectangle is calculated (step S104). Here, if the similarity is greater than or equal to a predetermined threshold, the target rectangle is assumed to match the feature data of that hierarchy. If it matches the feature data (step S105 / YES) and the hierarchy is the last hierarchy (step S106 / YES), it is determined that the rectangular area is a face, and information on the rectangular area is registered in the memory (step S108). ). If the hierarchy is not the final hierarchy (step S106 / NO), the hierarchy L is incremented by 1 (step S107), and the similarity with the feature data of the next hierarchy is calculated (step S104). If the feature data in the hierarchy is not matched (step S105 / NO), it is determined that the target rectangular area is not a face at that time.

  Here, it is determined whether or not the camera has moved (step S109). The camera movement is determined based on a change in the output value of the acceleration sensor 90. The output value of the acceleration sensor 90 is stored in a memory at a predetermined interval (here, 100 ms interval), and the difference from the sensor output at the previous motion determination is checked at the time of motion determination. When the sensor output difference exceeds a predetermined threshold (angle), it is determined that the camera has moved. At this time, the predetermined threshold used for the determination is different depending on the zoom position of the lens. When the zoom position is a wide angle, the change in the monitoring angle of view with respect to the angle change is relatively small. Therefore, the threshold is set to be large. It is assumed that the threshold is small. If it is determined that the camera has moved (step S109 / YES), there is a high possibility that the position of the face has changed from the currently processed monitoring image, so the current detection result is cleared (step S110). ) Re-acquire the face detection monitoring image and repeat the detection operation from the beginning.

  If it is determined that there is no motion in the camera (step S109 / NO), the face detection process is continued as it is. If it is determined whether or not the target rectangular area is a face, it is confirmed whether or not the entire face detection target area in the image has been scanned (step S111). If the area to be scanned still remains (step S111 / NO), the next rectangular area is shifted from the previous target rectangular area by a predetermined step width, for example, a pixel corresponding to 10% of the rectangular area being scanned. Determine (step S112).

  When it is determined whether or not a certain target rectangular area is a face, if the entire face detection target area has been scanned (YES in step S111), the face detection processing for the image ends, and face detection is performed. When the mode is continued, a new image is copied from the monitoring data, and face detection processing is similarly performed on the image (step S113 / YES). If the face detection mode is not continued (step S113 / NO), the process ends here.

[Second Embodiment]
FIG. 6 shows a flow of face detection processing corresponding to the second embodiment. When the camera operation mode is a mode for performing face detection processing, the face detection module copies one image from the monitoring image at a certain timing (step S201). The CPU 43 performs face detection processing on the copied image and determines whether or not there is a human face in the image.

  In the face detection process, the sum of the similarities between the feature data of a person's face stored in advance in the memory and the symmetric rectangular area in the image is calculated as the similarity between the target rectangular area and the feature data. To do. Here, the facial feature data and the similarity calculation method are the same as those in the first embodiment.

  First, the first target rectangular area is determined from the image (step S202). Here, it is assumed that the upper left edge of the image is the same as in the first embodiment. Also in the present embodiment, as in the first embodiment, as shown in FIG. 4, scanning is performed while gradually shifting the target rectangular area in order from the upper left end of the image to the lower right end.

  First, the similarity with the feature data is calculated for the target rectangular area determined in step S202 (step S203). If the degree of similarity is equal to or greater than a predetermined threshold, the target rectangular area matches the facial feature data. If the target rectangular area matches the feature data (step S204 / YES), information on the rectangular area is registered in the memory assuming that the target rectangular area is a face (step S205).

  Here, it is determined whether or not there is a motion in the monitoring image (step S206). The motion determination of the image is performed based on the difference of the average luminance value of each block by dividing the monitoring image into 32 × 32 blocks. The difference between the average luminance of each block in the monitoring image at the start of the face detection process and the average luminance of each block in the current monitoring image is taken, and it is determined that the block whose difference exceeds the threshold has motion. At this time, if the number of blocks determined to have motion exceeds a predetermined threshold, it is determined that the entire image has motion. If it is determined that there is a motion (step S206 / YES), there is a high possibility that the position of the face has changed from the currently processed monitoring image, so the current detection result is cleared (step S207), Re-acquire the face detection monitoring image and repeat the detection operation from the beginning.

  If it is determined that there is no movement in the image (step S206 / NO), the face detection process is continued as it is. If it is determined whether or not the target rectangular area is a face, and there remains an unscanned area in the face detection target area in the image (step S208 / NO), it is based on the similarity in the previous rectangular target area. The step width for determining the next target rectangular area is calculated (step S209). Here, 10% of the size of the face detection target rectangular area is set as the step width, and a rectangular area shifted by the step width from the previous face detection target rectangular area is set as the next detection target rectangular area.

  When it is determined whether or not a certain rectangular area is a face, if the entire face detection target area has been scanned (step S208 / YES), the face detection processing for the image is terminated, and the face When the detection mode is continued, a new image is copied from the monitoring data, and face detection processing is similarly performed on the image (step S210 / YES). If the face detection mode is not continued (step S210 / NO), the process ends here.

[Third Embodiment]
FIG. 7 shows a flow of face detection processing corresponding to the third embodiment. When the camera operation mode is a mode for performing face detection processing, the face detection module copies one image from the monitoring image at a certain timing (step S301). The CPU 43 performs face detection processing on the copied image and determines whether or not there is a human face in the image.

  In the face detection process, the sum of the similarities between the feature data of a person's face that is stored in advance in the memory and the symmetric rectangular area in the image is calculated as the similarity between the target rectangular area and the feature data. And Here, the facial feature data and the similarity calculation method are the same as those in the first embodiment.

  First, the first target rectangular area is determined from the image (step S302). Here, it is assumed that the upper left edge of the image is the same as in the first embodiment. Also in the present embodiment, as in the first embodiment, as shown in FIG. 4, scanning is performed while gradually shifting the target rectangular area in order from the upper left end of the image to the lower right end.

  First, the similarity with the feature data is calculated for the target rectangular area determined in step S302 (step S303). If the degree of similarity is equal to or greater than a predetermined threshold, the target rectangular area matches the facial feature data. If the target rectangular area matches the feature data, information on the rectangular area is registered in the memory assuming that the target rectangular area is a face (step S305).

  Here, it is determined whether or not there is movement in the monitoring image (steps S306 and S308). The motion determination of the image is performed based on the difference of the average luminance value of each block by dividing the monitoring image into 32 × 32 blocks. The difference between the average luminance of each block in the monitoring image at the start of the face detection process and the average luminance of each block in the current monitoring image is taken, and it is determined that the block whose difference exceeds the threshold has motion. At this time, the face detection process is continued, the face detection process is re-executed, or the process shifts to the tracking process, depending on the number of blocks determined to have motion. When the number of blocks determined to be moving is equal to or greater than the predetermined threshold value 1 (Th1) (step S306 / YES), there is a high possibility that the position of the face has changed significantly from the monitoring image currently being processed. Therefore, the current detection result is cleared (step S307), the monitoring image for face detection is re-acquired, and the detection operation is restarted from the beginning.

  If the number of blocks determined to be moving is less than the predetermined threshold 1 but greater than or equal to the predetermined threshold 2 (Th2 (Th1> Th2)) (step S306 / NO, step S308 / YES), It is highly possible that the face position has changed from the monitoring image being processed, but there is a high possibility that there is a face near the face position of the monitoring image being processed. Then, a new monitoring image is acquired (step S309), and tracking processing based on the current detection result is performed (step S310). In the present embodiment, the tracking process is based on the face detection result or the face tracking result in the previous cycle and is limited to a partial region around the face position in the entire cycle in the same manner as the face detection process. The processing is to detect the area.

  If the number of blocks determined to have motion is less than the predetermined threshold 2 (step S308 / NO), it is determined that there is no motion in the image, and the face detection process continues. If it is determined whether or not the target rectangular area is a face, and there remains an unscanned area in the face detection target area in the image (step S311 / NO), it is based on the similarity in the previous rectangular target area. The step width for determining the next target rectangular area is calculated (step S312). Here, 10% of the size of the face detection target rectangular area is set as the step width, and a rectangular area shifted by the step width from the previous face detection target rectangular area is set as the next detection target rectangular area.

  When it is determined whether or not a certain target rectangular area is a face, if the entire face detection target area has been scanned (step S311 / YES), the face detection process for the image is terminated, and face detection is performed. When the mode is continued, a new image is copied from the monitoring data, and face detection processing is similarly performed on the image (step S313 / YES). If the face detection mode is not continued (step S313 / NO), the process ends here.

  According to the above embodiment, it is determined that the angle of view of the camera has changed, and unnecessary calculation processing for the monitoring image with the old angle of view is stopped, so that the face display frame is prevented from being shifted from the actual face position. This makes it possible to detect an image with a new angle of view without wasting time.

  In addition, according to the above-described embodiment, optimal calculation processing is performed in accordance with the amount of change in the attitude of the imaging apparatus, so that it is possible to minimize the dead time required for face detection.

  Further, according to the above-described embodiment, by performing each process of AF, AE, and AWB based on the result of face detection, even when the person is not near the center of the image, the person can be easily focused. It is possible to achieve proper exposure.

  The above-described embodiment is a preferred embodiment of the present invention, and the scope of the present invention is not limited to the above-described embodiment alone, and various modifications are made without departing from the gist of the present invention. Implementation is possible.

  That is, the program executed by the imaging apparatus according to the present embodiment has a module configuration including the above-described units (view angle change detection unit and face detection unit), and is implemented using actual hardware. Realize the means. That is, when the computer (CPU) reads a program from a predetermined recording medium and executes it, the above-mentioned means are loaded onto the main storage device and generated.

  The program executed on the client PC in the present embodiment may be stored on a computer connected to a network such as the Internet and provided by being downloaded via the network. Further, the program may be provided or distributed via a network such as the Internet.

  The program is a file in an installable or executable format, such as a floppy (registered trademark) disk, hard disk, optical disk, magneto-optical disk, CD-ROM, CD-R, DVD, nonvolatile memory card, or the like. It may be configured to be provided by being recorded on a computer-readable recording medium. Further, the program may be provided by being incorporated in advance in a ROM or the like.

  In this case, the program code itself read from the recording medium or loaded and executed through the communication line realizes the functions of the above-described embodiments. And the recording medium which recorded the program code comprises this invention.

1 Power button 2 Release shutter button 3 Mode dial 4 Optical viewfinder 5 LCD
6 Flash 7 Shooting lens 11 Zoom / AF lens 12 Aperture and color filter 13 Mechanical shutter 20 Motor driver 30 Image sensor 31 Light receiving sensor 32 A / D converter 33 TG
40 Digital signal processing IC
41 Sensor I / F
42 memory controller 43 CPU
44 YUV converter 45 Compression / decompression circuit 46 Display output control 47 Resizing process 48 Media I / F
49 Acceleration sensor I / F
50 SDRAM
60 Operation section 70 ROM
80 memory card 90 acceleration sensor

JP 2007-306416 A

Claims (16)

Imaging means for inputting image data of the photographed subject;
An angle-of-view change detecting means for detecting that the angle of view of the image data has changed;
Face detection means for detecting a human face from a captured image using image data obtained by the imaging means;
Have
If a change in the angle of view is detected by the angle-of-view change detection unit during the face detection calculation process, the face detection unit temporarily stops the calculation process, newly acquires an image, and performs the face detection An imaging apparatus that performs arithmetic processing. It has posture change amount detection means for detecting the amount of change in posture of the imaging device itself,
The imaging apparatus according to claim 1, wherein the angle-of-view change detection unit determines whether or not there is a change in the angle of view of the image data based on a change amount detected by the posture change amount detection unit. It has posture change amount detection means for detecting the amount of change in posture of the imaging device itself,
The angle-of-view change detection means determines the presence or absence of a change in the angle of view of the image data based on the change amount detected by the posture change amount detection means and a predetermined threshold value,
The imaging apparatus according to claim 1, wherein the predetermined threshold value varies depending on a focal length of the imaging apparatus.   The imaging apparatus according to claim 1, wherein the angle-of-view change detection unit determines presence or absence of a change in angle of view of the image data based on a pixel luminance value difference for each divided region in the captured image. It has face tracking means for tracking the face with a small amount of calculation based on the previous face detection result,
5. The imaging apparatus according to claim 4, wherein a face detection process by the face detection unit or a face tracking process by the face tracking unit is executed based on a pixel luminance value difference for each divided region in the photographed image. . The face detection unit stops the face detection process during calculation and acquires a new image when a divided area determined to have movement based on the pixel luminance value difference exceeds a first threshold. To perform face detection processing,
The face tracking means is performing computation when a divided area determined to have movement based on the pixel luminance value difference exceeds a second threshold that is equal to or less than the first threshold and smaller than the threshold. Stop the face detection process, acquire a new image, perform the face tracking process,
The face detection unit continuously performs face detection processing during calculation when a divided region determined to have motion based on the pixel luminance value difference is equal to or less than the second threshold value. The imaging device according to claim 5.   7. At least one of automatic focus adjustment, automatic exposure adjustment, and automatic white balance is executed based on face area information obtained by the face detection means. The imaging device described in 1. An imaging step of acquiring image data of the photographed subject;
An angle-of-view change detecting step for detecting that the angle of view of the image data has changed;
Using the image data obtained in the imaging step, a face detection step of detecting a human face from the captured image;
Have
In the face detection step, when a change in the angle of view is detected by the angle-of-view change detection step during the face detection calculation process, the calculation process is temporarily stopped, and a new image is acquired to detect the face detection. A face detection processing method characterized by performing arithmetic processing. In a computer of an imaging apparatus having an imaging means for inputting image data of a photographed subject,
An angle-of-view change detection function for detecting that the angle of view of the image data has changed;
A face detection function for detecting a person's face from a photographed image using image data obtained by the imaging means;
Realized,
In the face detection function, if a view angle change is detected by the view angle change detection function during the face detection calculation process, the face detection function temporarily stops the calculation process, newly acquires an image, and performs the face detection function. A program characterized by performing arithmetic processing. Let the computer of the imaging device realize a posture change amount detection function for detecting the amount of change in posture of the imaging device itself,
The program according to claim 9, wherein the angle-of-view change detection function determines whether or not there is a change in the angle of view of the image data based on a change amount detected by the posture change amount detection function. Let the computer of the imaging device realize a posture change amount detection function for detecting the amount of change in posture of the imaging device itself,
The angle-of-view change detection function determines the presence or absence of a change in the angle of view of the image data based on the change amount detected by the posture change amount detection function and a predetermined threshold,
The program according to claim 9 or 10, wherein the predetermined threshold value varies depending on a focal length of the imaging apparatus.   The program according to claim 9, wherein the angle-of-view change detection function determines whether there is a change in the angle of view of the image data based on a pixel luminance value difference for each divided region in the captured image. The computer of the imaging device realizes a face tracking function for tracking a face with a small amount of calculation based on the previous face detection result,
The program according to claim 12, wherein a face detection process by the face detection function or a face tracking process by the face tracking function is executed based on a pixel luminance value difference for each divided region in the captured image. The face detection function cancels the face detection process during calculation and acquires a new image when a divided area determined to have motion based on the pixel luminance value difference exceeds a first threshold. To perform face detection processing,
The face tracking function is performing an operation when a divided region determined to have movement based on the pixel luminance value difference exceeds a second threshold value that is equal to or less than the first threshold value and smaller than the threshold value. Stop the face detection process, acquire a new image, perform the face tracking process,
The face detection function continuously performs face detection processing during calculation when a divided region determined to have movement based on the pixel luminance value difference is equal to or less than the second threshold value. The program according to claim 13.   The computer of the imaging apparatus is configured to realize at least one of an automatic focus adjustment function, an automatic exposure adjustment function, and an automatic white balance function based on information on a face area obtained by the face detection function. Item 15. The program according to any one of Items 9 to 14.   A recording medium that records the program according to any one of claims 9 to 15 and is readable by a computer.

Images