JP2014056169A - Focus adjustment device and focus adjustment method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a focus adjustment device having high AF accuracy for a face, even when an object does not face the front and to provide a focus adjustment method.SOLUTION: The focus adjustment device includes a face detection part detecting a face area including the face of the object in an object image, based on image data generated by imaging the object image formed by a photographic optical system, an AF part setting an AF area in a photographing screen and executing an AF operation, based on the image data on the AF area, a rotation angle calculation part calculating the rotation angle of the face, based on the image data of the face area detected by the face detection part, and an AF area position correction part correcting the position of the AF area, based on the rotation angle calculated by the rotation angle calculation part.

Description

  The present invention relates to a focus adjustment device and a focus adjustment method for performing a focus adjustment operation on a subject.

  2. Description of the Related Art Conventionally, a technique of face detection processing for detecting a human face image included in image data acquired by an image sensor is known. There is also known an imaging apparatus that performs focusing control so that the face image detected by the face detection process is in focus.

  In Patent Document 1, face detection processing is performed to detect a person's face based on image data, the face area is detected, and the AF area is changed according to the size of the face area, and an AF operation is performed. It is disclosed. By such processing, an AF error caused by the background included in the AF area is suppressed.

JP 2009-098317 A

  However, in the case of Patent Document 1, when the subject is not facing the front, for example, when the subject is a side profile, if the size of the AF area is reduced by the same method, the background is included in the AF area, and the AF accuracy decreases. The problem remains.

  The present invention has been made in view of the above circumstances, and an object thereof is to provide a focus adjustment device and a focus adjustment method with high AF accuracy with respect to a face even when a subject is not facing the front. To do.

  In order to achieve the above object, a focus adjustment apparatus according to a first aspect of the present invention is based on image data generated by imaging a subject image formed by a photographing optical system, and A face detection unit that detects a face region including the face of the subject, an AF control unit that sets an AF area in a shooting screen and executes an AF operation based on image data related to the AF area, and the face detection unit A rotation angle calculation unit that calculates the rotation angle of the face based on the image data of the detected face area, and an AF area position correction unit that corrects the position of the AF area based on the rotation angle calculated by the rotation angle calculation unit And.

  In the focus adjustment apparatus according to a second aspect of the present invention, in the first aspect, the rotation angle calculation unit uses the first rotation angle with the face upright direction as the axis and the face-to-face direction as the axis. A second rotation angle is calculated.

  In the focus adjustment apparatus according to a third aspect, in the first aspect, the AF area position correction unit changes the area of the AF area based on the rotation angle calculated by the rotation angle calculation unit.

  According to a fourth aspect of the present invention, there is provided a focus adjustment method, wherein a face region including a face of the subject is included in the subject image based on image data generated by capturing a subject image formed by a photographing optical system. Detecting, calculating a rotation angle of the face based on the detected image data of the face area, correcting the position of the AF area based on the calculated rotation angle, and an image corresponding to the corrected AF area A focus adjustment operation is performed based on the data.

  According to the present invention, it is possible to provide a focus adjustment device and a focus adjustment method with high AF accuracy for a face even when the subject is not facing the front.

It is a figure which shows the structure as an example of the imaging device provided with the focus adjustment apparatus which concerns on embodiment of this invention. 4 is a flowchart illustrating a shooting operation of the imaging apparatus according to the embodiment of the present invention. It is a figure for demonstrating the detection method of a face area. It is a flowchart shown about the position correction process of AF area which concerns on embodiment of this invention. It is a figure for demonstrating the rotation angle and left-right inclination of a face. It is a figure for demonstrating the correction value according to the rotation angle and left-right inclination of a face. It is a figure for demonstrating in detail about the rotation angle of a face. It is a figure for demonstrating in detail about the left-right inclination of a face. It is a figure for demonstrating the operation | movement which correct | amends the position of AF area according to an example of the condition of a face. It is a figure for demonstrating the operation | movement which correct | amends the position of AF area according to the other example of the condition of a face.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. FIG. 1 is a diagram illustrating a configuration as an example of an imaging apparatus including a focus adjustment apparatus according to each embodiment of the present invention. An imaging apparatus 100 illustrated in FIG. 1 includes a photographing optical system 102, a focus adjustment mechanism 104, a diaphragm 106, a diaphragm driving mechanism 108, a shutter 110, a shutter driving mechanism 112, an imaging element 114, and an imaging element interface ( IF) circuit 116, RAM 118, display element 120, display element drive circuit 122, touch panel 124, touch panel drive circuit 126, recording medium 128, system controller 130, operation unit 132, and ROM 134. Have.

  The photographing optical system 102 is an optical system for condensing a light flux F from a subject (not shown) on the light receiving surface of the image sensor 114. The photographing optical system 102 has a plurality of lenses such as a focus lens. The focus adjustment mechanism 104 includes a motor and its drive circuit. The focus adjustment mechanism 104 drives the focus lens in the photographing optical system 102 in the optical axis direction (the one-dot chain line direction in the drawing) under the control of the CPU 1301 in the system controller 130.

  The diaphragm 106 is configured to be openable and closable, and adjusts the amount of the light beam F incident on the image sensor 114 via the photographing optical system 102. The aperture drive mechanism 108 has a drive mechanism for driving the aperture 106. The diaphragm driving mechanism 108 drives the diaphragm 106 according to the control of the CPU 1301 in the system controller 130.

  The shutter 110 is configured to place the light receiving surface of the image sensor 114 in a light shielding state or an exposure state. The exposure time of the image sensor 114 is adjusted by the shutter 110. The shutter drive mechanism 112 has a drive mechanism for driving the shutter 110, and drives the shutter 110 under the control of the CPU 1301 in the system controller 130.

  The imaging element 114 has a light receiving surface on which a light flux F from a subject condensed through the photographing optical system 102 is imaged. The light receiving surface of the image sensor 114 is configured by two-dimensionally arranging a plurality of pixels, and a color filter is provided on the light incident side of the light receiving surface. Such an image sensor 114 converts an image (subject image) corresponding to the light beam F formed on the light receiving surface into an electrical signal (hereinafter referred to as an image signal) corresponding to the light amount. Here, as the imaging device 114, imaging devices having various configurations such as a CCD system and a CMOS system are known. Various color filters such as a Bayer array are also known. In the present embodiment, the configuration of the image sensor 114 is not limited to a specific configuration, and image sensors having various configurations can be used.

  The image sensor IF circuit 116 drives the image sensor 114 under the control of the CPU 1301 in the system controller 130. The image sensor IF circuit 116 reads the image signal obtained by the image sensor 114 according to the control of the CPU 1301 in the system controller 130, and performs CDS (correlated double sampling) processing or AGC (automatic operation) on the read image signal. Analog processing such as gain control processing is performed. Further, the image sensor IF circuit 116 converts the analog processed image signal into a digital signal (hereinafter referred to as image data).

  A RAM 118 serving as a storage unit is, for example, an SDRAM and has a work area. The work area is a storage area provided in the RAM 118 for temporarily storing data generated in each part of the imaging apparatus 100 such as image data obtained by the imaging element IF circuit 116.

  The display element 120 is a liquid crystal display (LCD), for example, and displays various images such as an image for live view and an image recorded on the recording medium 128. The display element driving circuit 122 drives the display element 120 based on the image data input from the CPU 1301 of the system controller 130 and causes the display element 120 to display an image.

  The touch panel 124 is integrally formed on the display screen of the display element 120, and detects a contact position of a user's finger or the like on the display screen. The touch panel drive circuit 126 drives the touch panel 124 and outputs a contact detection signal from the touch panel 124 to the CPU 1301 of the system controller 130. The CPU 1301 detects a contact operation on the display screen of the user from the contact detection signal, and executes processing according to the contact operation.

  The recording medium 128 is a memory card, for example, and records an image file obtained by a shooting operation. The image file is a file configured by adding a predetermined header to image data. In the header, data indicating shooting conditions is recorded as tag data.

  The system controller 130 is a control circuit for controlling the operation of the imaging apparatus 100, and includes a CPU 1301, an AF control circuit 1302, an AE control circuit 1303, an image processing circuit 1304, a face detection circuit 1305, and a memory control circuit 1306. And have.

  The CPU 1301 has a function as a control unit. Each block outside the system controller 130 such as the focus adjustment mechanism 104, the aperture driving mechanism 108, the shutter driving mechanism 112, the display element driving circuit 122, and the touch panel driving circuit 126, and the system The operation of each control circuit in the controller 130 is controlled.

  The AF control circuit 1302 controls contrast AF processing. Specifically, the AF control circuit 1302 extracts high-frequency components of image data corresponding to a predetermined AF area from the image data obtained by the image sensor IF circuit 116, and integrates the extracted high-frequency components. The focus evaluation value for AF is acquired. The CPU 1301 controls the focus adjustment mechanism 104 to bring the focus lens into focus while evaluating the contrast of the image data according to the focus evaluation value. The CPU 1301 and the AF control circuit 1302 constitute an AF control unit.

  The AE control circuit 1303 controls the AE operation. Specifically, the AE control circuit 1303 calculates subject luminance using the image data obtained by the image sensor IF circuit 116. The CPU 1301 calculates the aperture amount (aperture value) of the aperture 106 during exposure, the opening time of the shutter 110 (shutter speed value), the image sensor sensitivity, the ISO sensitivity, and the like according to the subject brightness.

  The image processing circuit 1304 performs various image processing on the image data. This image processing includes color correction processing, gamma (γ) correction processing, compression processing, and the like. The image processing circuit 1304 also performs decompression processing on the compressed image data.

  The face detection circuit 1305 detects a face area by a face detection algorithm using a known pattern recognition technique. A feature amount corresponding to the eyes and nose is extracted from the region that is a face candidate by detecting the contour shape and skin color region, and the extracted feature amount is compared with the feature amounts of a plurality of face template images. It is determined whether the candidate is a person's face. When the face detection circuit 1305 discriminates and detects the face, it calculates the position (coordinates) of the face area, the size of the face area, the rotation angle of the face, and the reliability of face detection, and outputs them to the CPU 1301. The face detection circuit 1305 also has a function of detecting an eye (pupil) area, calculates the position (coordinates) and size of the eye area, and outputs the calculated position to the CPU 1301. The face detection circuit 1305 constitutes a face detection unit and a rotation angle detection unit.

  The CPU 1301 constitutes an AF area position correction unit that corrects the position of the AF area according to the face area data and the face rotation angle data output from the face detection circuit 1305.

  The memory control circuit 1306 is an interface that performs control for the CPU 1301 and the like to access the RAM 118, the recording medium 128, and the ROM 134.

  The operation unit 132 is various operation members operated by the user. The operation unit 132 includes, for example, a release button, a moving image button, a mode dial, a selection key, a power button, and the like.

  The release button button has a 1st release switch and a 2nd release switch. The 1st release switch is a switch that is turned on when the user half-presses the release button. When the 1st release switch is turned on, a shooting preparation operation such as AF processing is performed. The 2nd release switch is a switch that is turned on when the user fully presses the release button. When the 2nd release switch is turned on, an exposure operation for still image shooting is performed.

  The moving image button is an operation member for instructing the start or end of moving image shooting. When the moving image button is pressed by the user, the moving image shooting process is started. Further, when the moving image button is pressed during the moving image shooting process, the moving image shooting process is ended.

  The mode dial is an operation member for selecting shooting settings of the imaging apparatus. In the present embodiment, for example, a still image shooting mode and a moving image shooting mode can be selected as shooting settings of the imaging apparatus. The still image shooting mode is a shooting setting for shooting a still image. The moving image shooting mode is a shooting setting for shooting a moving image.

  The selection key is an operation member for selecting or determining an item on the menu screen, for example. When the selection key is operated by the user, an item is selected or determined on the menu screen.

  The power button is an operation member for turning on or off the power of the imaging apparatus. When the power button is operated by the user, the imaging apparatus 100 is activated and becomes operable. If the power button is operated while the imaging apparatus is activated, the imaging apparatus 100 enters a power saving standby state.

  The ROM 134 is, for example, a Flash ROM, and stores program codes for the CPU 1301 to execute various processes. In addition, the ROM 134 stores various control parameters such as control parameters necessary for the operation of the photographing optical system 102, the aperture 106, the image sensor 114, and the like, and control parameters necessary for image processing in the image processing circuit 1304. ing. Further, the ROM 134 stores data for correcting the position of the AF area.

  Next, the operation of the imaging apparatus according to the present embodiment will be described. FIG. 2 is a flowchart illustrating the shooting operation of the imaging apparatus 100. The CPU 1301 reads the necessary program code from the ROM 134 and controls the operation of FIG.

  In S100, the CPU 1301 starts a live view operation. As a live view operation, the CPU 1301 controls the shutter drive mechanism 112 to release the shutter 110, and then the CPU 1301 controls the image sensor IF circuit 116 to start imaging by the image sensor 114. Thereafter, the CPU 1301 inputs the image data stored in the work area of the RAM 118 as a result of imaging by the imaging device 114 to the image processing circuit 1304 and performs image processing for live view display. Subsequently, the CPU 1301 inputs image data that has been subjected to image processing for live view display to the display element driving circuit 122 and causes the display element 120 to display an image. By repeatedly executing such a display operation, the subject image is displayed as a moving image. By this moving image display, the user can observe the subject.

  In step S102, the CPU 1301 instructs the face detection circuit 1305 to execute face detection processing. The face detection circuit 1305 outputs to the CPU 1301 information indicating whether or not face detection has been performed by performing face detection processing, and various information related to the detected face area, for example, information such as the position and size of the face area. To do.

  In step S <b> 104, the CPU 1301 determines whether face detection has been performed based on information output from the face detection circuit 1305. If the face is detected, the process proceeds to S106. If the face is not detected, the process proceeds to S108.

In step S106, the CPU 1301 instructs the display element driving circuit 122 to display a display frame indicating the latest face area in which the face is detected. The display element driving circuit 122 performs display indicating the face area by the display element 120 in accordance with an instruction from the CPU 1301. In S108, the CPU 1301 determines whether or not the 1st release switch is turned on. If it is determined in S108 that the first release switch is not turned on, the CCPU 1301 returns the process to S102 and executes the face detection process. Thus, until the 1st release switch is turned on, the CPU 1301 continues the live view operation, updates the detected face area, and displays the face area.

  If it is determined in S108 that the 1st release switch is turned on, in S110, the CPU 1301 determines whether or not the face is finally detected. Then, the CPU 1301 advances the process to S112 when the face is not detected, and advances to S114 when the face is detected.

  In step S112, the CPU 1301 sets a predetermined AF area. For example, an AF area corresponding to the multi AF mode in which a plurality of AF areas are arranged in the shooting screen is set. Then, the process proceeds to S120.

  In step S114, the CPU 1301 determines whether the face detection circuit 1305 can detect the pupil. Then, the CPU 1301 advances the process to S116 when the pupil is not detected, and advances to S118 when the pupil is detected. In the pupil detection process, when the face is large to some extent, the amount of image data corresponding to the pupil portion is large, so that the pupil can be detected with high accuracy. However, as the face becomes smaller, the amount of image data corresponding to the pupil portion becomes smaller and cannot be detected.

  In step S116, the CPU 1301 executes AF area position correction, which will be described later, and sets a corrected AF area. In step S118, the CPU 1301 sets an AF area corresponding to the position of the pupil region based on the result of pupil detection by the face detection circuit 1305. Then, the process proceeds to S120.

  In S120, the CPU 1301 executes AF processing. In the AF process, the focus lens is driven to the in-focus position by scan driving. In the scan drive, the CPU 1301 controls the focus adjustment mechanism 104 to drive the focus lens in one direction within a predetermined scan range, and the focus is sequentially calculated by the AF control circuit 1302 based on the image data in the AF area. Evaluate the evaluation value. Then, the CPU 1301 stops the drive of the focus lens at the position of the focus lens where the focus evaluation value is maximum, that is, the image contrast is maximum, as a result of the evaluation of the focus evaluation value.

  Further, the CPU 1301 performs the AE process in parallel with the execution of the AF process. In the AE process, the CPU 1301 sets the aperture of the aperture 106 during the main exposure (the appropriate exposure amount) so that the luminance of the subject located in the AF area calculated by the AE control circuit 1303 is a predetermined appropriate amount (appropriate exposure amount). Aperture value) and the opening time of the shutter 110 (shutter speed value) are calculated.

  In step S122, the CPU 1301 determines whether the 2nd release switch is turned on, and waits until the 2nd release switch is turned on. When it is determined that the 2nd release switch is turned on, an exposure operation is performed in S123.

  In the exposure operation, the CPU 1301 controls the shutter drive mechanism 112 to close the shutter 110 and controls the aperture drive mechanism 108 to narrow the aperture 106 to the previously calculated aperture value. Subsequently, the CPU 1301 controls the shutter driving mechanism 112 to perform an imaging operation by the imaging element 114 while opening the shutter 110 for the previously calculated opening time. Thereafter, the CPU 1301 causes the image processing circuit 1304 to process the image data obtained by the imaging operation of the imaging element 114 to obtain still image data. In step S <b> 124, the CPU 1301 performs processing for recording still image data on the recording medium 128. Thereafter, the CPU 1301 ends the process shown in FIG.

  Next, a method for setting a face area based on the result of face detection executed by the face detection circuit 1305 will be described. FIG. 3 shows setting of the face detection area for the face F.

  First, setting of the width of the face area will be described. Both eyes (pupils) are detected from the image data of the face F shown in FIG. 3, and the interval between the eyes is set to Le. Here, based on statistics, it is generally assumed that the face width is a predetermined multiple of the distance between the eyes. Based on this assumption, N × Le obtained by multiplying the distance Le between both eyes by a predetermined ratio N and equally distributed from the center of both eyes is set as the width Lf1 of the face area Fa.

  Next, setting of the length direction of the face area will be described. Based on statistics, it is generally assumed that the eye position in the face length direction is a position obtained by dividing the face length by a predetermined ratio. Based on this assumption, a range of a predetermined distance in the face length direction is set as a face area with the eye position in the face length direction as a reference.

  Both eyes are detected from the image data of the face F shown in FIG. 3, and the position of the eyes in the length direction of the face is determined. Then, assuming that the ratio of the position of the eyes that divides the length of a general face based on statistics in the length direction is L: M, the region range in the head (up) direction in the length direction of the face with respect to the eyes Lte (FLE × L / (L + M)), a region range Lbe (FLE × M / (L + M)) in the chin (downward) direction of the eye is set (the length of the face is assumed to be FLE).

  Such face area setting processing is executed by the face detection circuit 1305 as part of face detection processing (FIG. 2: S102), and the center coordinates Fc of the face area Fa and the size S (Lf1, Lf2, area) of the face area Etc.) is output to the CPU 1301.

  Next, the AF area position correction process shown in S116 of FIG. 2 will be described with reference to FIG. In step S <b> 200, the CPU 1301 acquires the latest face area information from the face detection circuit 1305. In addition to the size and position of the face area, the face area information includes a face rotation angle, face left-right tilt information, and the like. In the face detection process, the face detection circuit 1305 calculates face rotation angle and face left-right tilt information by a known technique, and outputs the information to the CPU 1301.

  In step S <b> 202, the CPU 1301 selects a correction value area (R <b> 1 to R <b> 4 described later) based on the face rotation angle included in the face area information.

  In S204, the CPU 1301 corrects correction values (X-direction offset XO and Y-direction offset YO described later) in the correction value area (R1-R4) selected in S202 based on the left / right tilt information of the face included in the face area information. ) Is selected.

  In step S206, the CPU 1301 corrects the position of the AF area based on the selected correction value. Next, detailed processing of AF area position correction will be described.

  FIG. 5 is a diagram for explaining a face rotation angle (first rotation angle) and a horizontal tilt angle (second rotation angle), which are input conditions for correcting the position of the AF area.

  As shown in FIG. 5A, the rotation angle of the face is an angle at which the face rotates about the axis F-R in the front-rear direction of the head H of the person. The axis F-R passes vertically through the center of the face, for example, the center of the nose, when the face is a plane.

  Further, as shown in FIG. 5B, the left-right inclination of the face is an angle at which the face rotates about the vertical axis TB of the person's head H. The axis TB is set so as to pass through the center when the head is viewed from above.

  FIG. 6 is a table showing correction values relating to the correction of the position of the AF area according to the rotation angle (first rotation angle) of the person's face and the left / right tilt information (second rotation angle) of the face. The table data of FIG. 6 is stored in the ROM 134 as a part of the control parameter.

  The CPU 1301 reads the table data shown in FIG. 6 from the ROM 134 and corrects the position of the AF area. In FIG. 6, the correction values are an X-direction offset XO and a Y-direction offset YO, and each unit is%.

  In FIG. 6, the rotation angle θ of the face is divided into four regions R1 to R4, and the left / right inclination δ of the face is divided into three regions, left, front, and right. Then, correction values (X direction offset XO, Y direction offset YO) are set according to these combinations.

  FIG. 7 is a diagram for explaining division (R1 to R4) of the face rotation angle θ (first rotation angle). The face rotation angle θ is the rotation angle of the face with the front-rear direction of the face as an axis as shown in FIG. 5A, but the face rotation in a state where the person is upright as shown in FIG. The angle is 0. For example, the face rotation angle when the direction TB changes to the direction T′-B ′ due to the rotation of the face is θ. In FIG. 7, the rotation angle θ is a candidate for selecting a correction value included in the region R2 and corresponding to the region R2.

  FIG. 8 is a diagram for explaining division (left, front, right) of the left-right inclination δ (second rotation angle) of the face. As shown in FIG. 5B, the left / right tilt information of the face is a rotation angle of the face with the vertical direction of the face as an axis. As shown in FIG. The left / right tilt information in the direction of For example, when the direction FB is changed to the direction F′-B ′ by the left-right rotation of the face, the left-right inclination of the face is δ. Note that the right / left tilt information δ of the face is defined by the orientation of the face as viewed from the photographer, and in FIG. 8, the left / right tilt information δ is included in the leftward area. In this case, the correction value corresponding to the leftward area is selected.

In this way, the correction value shown in FIG. 6 is selected according to the combination of the face rotation angle θ and the left / right inclination δ of the face. Next, in order to correct the position of the AF area using the selected correction value. The AF area position correction formula will be described. Expressions (1) and (2) below are used to correct the AF area position.

Correction AF area position_X = face position center_X + (face size S × correction coefficient Hx × X direction offset XO) (1)
Correction AF area position_Y = face position center_Y + (face size S × correction coefficient Hy × Y direction offset YO) (2)
The face position center_X, the face position center_Y, and the face size S indicating the center position of the face area are data calculated by the face detection circuit 1305 and output to the CPU 1301.

  The corrected AF area position_X and the corrected AF area position_Y indicate the corrected position of the center of the AF area, and are calculated by the CPU 1301 using the above formulas (1) and (2).

  The CPU 1301 sets an AF area corrected around the corrected AF area position_X and the corrected AF area position_Y. The size of the corrected AF area is a size obtained by reducing the size (area) of the normal AF area by a predetermined ratio. The predetermined ratio may be 50%, for example. This is to prevent the AF area from being set so as to protrude from the face area, including the background, and reducing the focusing accuracy due to the influence of the background.

  The correction coefficients Hx and Hy are coefficients for finely adjusting the correction amount, and are set by the CPU 1301 according to the reliability of face detection, the reliability of angle detection output from the face detection circuit 1305, and other conditions. . The correction coefficients Hx and Hy are normally set to 1.

  FIG. 9 shows the setting of the AF area when the AF area position correction is performed together with the setting of the AF area when the AF area correction process is not performed, and shows the difference. In FIG. 9, AFEC ′ is the center of the face position indicated by the face position center_X and the face position center_Y detected by the face detection circuit 1305 with respect to the head H of the person whose face is to be detected. is there. AFE ′ indicates a face area detected by the face detection circuit 1305. As described above, the face area detection process has such a positional relationship because the algorithm is such that the center of the face area is set approximately in the vicinity of the person's nose.

  Here, when the AF processing is executed with the face area AFE ′ as the AF area, the face is inclined at an angle to the right for the photographer, so the AF area is set to a range that protrudes beyond the actual face. The background is included. For this reason, there arises a problem that the AF accuracy is lowered under the influence of the background.

  In this embodiment, the position of the AF area is corrected as follows. In the case of FIG. 9, since the face is upright, the rotation angle of the face is “0 °”, the face is directed rightward for the photographer, and the right / left tilt information of the face is “right”. . The face detection circuit 1305 determines the position of the AF area based on the rotation angle (0 °) of the face, the tilt information (right) of the left / right direction of the face, the table data of FIG. 6 and the equations (1) and (2). Make corrections.

  Specifically, when the table data of FIG. 6 is applied, since the face rotation angle θ = 0 is the region R1 and the face right / left tilt information δ = right, the X direction offset XO = −20% and the Y direction offset YO. = 15%. Specifically, when applied to the equations (1) and (2), the correction coefficients Hx and Hy are set to 1 as follows.

Correction AF area position_X = face position center_X + face size S × (−20%),
Corrected AF area position_Y = face position center_Y + face size S × 15%
Then, the face position center AFEC ′ before correction is changed to the face position center AFEC corrected by the face detection circuit 1305 with respect to the head H of the person subjected to face detection. Further, the face area AFE ′ before correction is changed to the face area AFE corrected by the face detection circuit 1305. Then, the CPU 1301 sets the corrected face area AFE as an AF area so that the AF area does not include a background, and can perform highly accurate AF without being affected by the background.

  Next, an example of a face situation different from FIG. 9 will be described with reference to FIG. In FIG. 10, the rotation angle of the face is 90 °, and the face is facing the right direction for the photographer. For the head H of the person who is the target of face detection, AFEC ′ is the center of the face position detected by the face detection circuit 1305, and AFE ′ indicates the face area detected by the face detection circuit 1305. . At this time, when the AF process is executed with the face area AFE as the AF area, the AF area is set in a range that protrudes beyond the actual face, and there is a problem that the background is included and the AF accuracy is lowered.

  In FIG. 10, the rotation angle of the face is “90 °”, the face is facing the right direction for the photographer, and the right / left tilt information of the face is “right”. The face detection circuit 1305 determines the position of the AF area based on the rotation angle (90 °) of the face, the tilt information (right) of the horizontal direction of the face, the table data in FIG. 6 and the equations (1) and (2). Make corrections.

  Specifically, when the table data of FIG. 6 is applied, since the face rotation angle θ = 90 ° is the region R2, and the right / left tilt information δ = right, the X direction offset = 15 and the Y direction offset = 20. Become. Specifically, when applied to the equations (1) and (2), the correction coefficients Hx and Hy are set to 1 as follows.

Correction AF area position_X = face position center_X + face size S × 15%,
Corrected AF area position_Y = face position center_Y + face size S × 20%
The face position center AFEC ′ before correction is changed to the face position center AFEC with respect to the head H of the person subjected to face detection, and the face area AFE ′ before correction is changed to the corrected face area AFE. Is done. Then, the CPU 1301 can set the corrected face area AFE as an AF area so that the AF area does not include a background, and can perform highly accurate AF without being affected by the background. is there.

  In the table data of FIG. 6, the X-direction offset XO and the Y-direction offset YO are set to correction values having different absolute values when the face is tilted in the left-right direction and has high contrast such as a person's ear or sideburns. This is to make it easy to detect the portion.

In the region R1 of FIG. 6 where the face is upright or close to it, the X-direction offset XO is a numerical value with a larger absolute value. Further, in the region R2 in FIG. 6 where the face is rotated 90 degrees from erect or close to it, the Y-direction offset YO is a numerical value having a larger absolute value.
<Modification>
In this embodiment, the face rotation angle is divided into four areas, and the right / left tilt information of the face is divided into three areas and the correction value is set. However, the number of divisions is not limited to this. May be increased or decreased. For example, the correction value can be set by dividing the face rotation angle into 8 finer areas, and the left / right tilt of the face can be set to “left”, “front”, “right”, The correction value may be set by adding “oblique rightward” and dividing into five regions. Alternatively, a predetermined arithmetic expression may be stored instead of the table data shown in FIG. 6, and the correction value may be calculated. The predetermined arithmetic expression may be stored in the ROM 134.

  Further, the correction coefficients of the equations (1) and (2) may be modified as follows.

  In the case of “slanting leftward” or “slanting rightward” where the tilt angle of the horizontal tilt is shallow, the correction amount may be set to 0.8 and the correction amount may be reduced. Further, in the case of “leftward” or “rightward” where the left / right inclination is deep, the correction amount may be increased by setting the correction coefficient to 1.2. When the face rotation angle is oblique, such as 45 ° or 135 °, the correction amount may be reduced by setting the correction coefficient to 0.7.

  As described above, the CPU 1301 sets the corrected AF area centered on the corrected AF area position_X and the corrected AF area position_Y to 50% of the size (area) of the normal AF area. Here, the size of the correction AF area may be changed according to the left / right tilt information of the face. For example, when the right / left tilt information δ of the face is in a predetermined angle range including the front, the same area as the normal AF area may be used, and in other angle ranges, the area may be 50% of the area of the normal AF area. . As a result, when the face that does not easily include the background is in the front (or in a situation close to the front), the AF area is set to be larger so that highly accurate AF can be performed even when the contrast is low.

  Furthermore, the reliability of the face angle information calculated by the face detection circuit 1305 may be determined, and when the angle information reliability is low, the correction amount may be reduced by setting the correction coefficient to 0.6. When the reliability of face detection or angle detection is low, overcorrection can be prevented.

  As described above, the face rotation angle (first rotation angle) and the left / right tilt information (second rotation angle) of the face are detected, and the correction value is determined based on the face rotation angle and left / right tilt information. By correcting the position of the AF area, it is possible to perform focus adjustment with high AF accuracy on the face even when the subject is not facing the front.

  Although the present invention has been described above based on the embodiments, the present invention is not limited to the above-described embodiments, and various modifications and applications are naturally possible within the scope of the gist of the present invention. Further, in the description of the operation described above, the operation is described using “first”, “next”, and the like for convenience. However, it does not mean that the operation is essential in this order.

  Further, the above-described embodiments include various stages of the invention, and various inventions can be extracted by appropriately combining a plurality of disclosed constituent elements. For example, even if some configuration requirements are deleted from all the configuration requirements shown in the embodiment, the above-described problem can be solved, and this configuration requirement is deleted when the above-described effects can be obtained. The configuration can also be extracted as an invention.

  DESCRIPTION OF SYMBOLS 100 ... Imaging device, 102 ... Imaging optical system, 104 ... Focus adjustment mechanism, 106 ... Aperture, 108 ... Aperture drive mechanism, 110 ... Shutter, 112 ... Shutter drive mechanism, 114 ... Imaging element, 116 ... Imaging element interface (IF) Circuit: 118 ... RAM, 120 ... Display element, 122 ... Display element drive circuit, 124 ... Touch panel, 126 ... Touch panel drive circuit, 128 ... Recording medium, 130 ... System controller, 132 ... Operation unit, 134 ... ROM, 1301 ... CPU DESCRIPTION OF SYMBOLS 1302 ... AF control circuit, 1303 ... AE control circuit, 1304 ... Image processing circuit, 1305 ... Face detection circuit, 1306 ... Memory control circuit

Claims (4)

A face detection unit that detects a face area including the face of the subject in the subject image based on image data generated by capturing a subject image formed by the photographing optical system;
An AF control unit that sets an AF area in the shooting screen and executes an AF operation based on image data related to the AF area;
A rotation angle calculation unit for calculating a rotation angle of the face based on the image data of the face area detected by the face detection unit;
An AF area position correction unit that corrects the position of the AF area based on the rotation angle calculated by the rotation angle calculation unit. The rotation angle calculation unit calculates a first rotation angle with the face upright direction as an axis and a second rotation angle with the face facing direction as an axis. Focusing device. The focus adjustment apparatus according to claim 1, wherein the AF area position correction unit changes the area of the AF area based on the rotation angle calculated by the rotation angle calculation unit. Based on image data generated by capturing a subject image formed by a photographing optical system, a face region including the subject's face in the subject image is detected;
Calculating a rotation angle of the face based on the detected image data of the face area;
Correct the position of the AF area based on the calculated rotation angle,
A focus adjustment method, wherein a focus adjustment operation is performed based on image data corresponding to the corrected AF area.

Images