JP5213639B2 - Image processing device - Google Patents

Description

  The present invention relates to an image processing apparatus, and more particularly to an image processing apparatus that is applied to a digital camera and processes an object scene image captured on an imaging surface.

An example of this type of device is disclosed in Patent Document 1. According to this background art, a plurality of face discrimination areas each having a plurality of different sizes are prepared. These face discrimination regions move the imaging surface in the raster scanning direction in order to detect a face image from the object scene image. When a partial image belonging to the face determination area is determined as a face image, the size and position of the face determination area at this time are described as face information. Imaging parameters such as focus and exposure amount are adjusted with reference to the face position information described in this way.
JP 2007-259423 A

  However, in the background art, the detection result of the face image is not reflected on the angle of view when the object scene image is reproduced, and the reproducibility of the object scene image is limited.

  Therefore, a main object of the present invention is to provide an image processing apparatus capable of improving the reproducibility of an object scene image.

The image processing apparatus according to the present invention (10:. Reference numerals corresponding to the same applies hereinafter in the examples), the capturing means for capturing an electronic image output from the imaging means (16) (30, 32), electrons captured by the capture means Detection means (S31 to S63, S89) for detecting a set of one or more specific object images from an image, and a set in which a cut-out area having a size based on the size of the set detected by the detection means is detected by the detection means Definition means (S105 to S107) for defining the covering position, and changing means (S109 to S107) for changing the size and / or position of the clipping area when the clipping area defined by the defining means protrudes from the electronic image captured by the capturing means. S111), and a part of the electronic image belonging to the cut-out area defined by the defining unit and modified as necessary by the changing unit. A clipping means (S1, S25, S27) for clipping from the electronic image captured by the stage is provided.

  Preferably, the detection means includes frame definition means (S89) for defining a frame that contacts the specific object images forming the set, and the definition means includes the frame defined by the frame definition means and the number of specific object images forming the set. Determining means (S105) for determining the size and position of the cut-out area based on

Preferably, the image display device further includes display means (36, 38) for displaying the electronic image cut out by the cut-out means, the imaging means repeatedly outputs the electronic image , and the capturing means repeatedly captures the electronic image .

  More preferably, it further comprises initial setting means (S81) for initial setting of the cutout area, and the definition means executes definition processing in response to a default instruction after the initial setting by the initial setting means.

Preferably, adjustment means (S91, S9, S15, S17, S23, S75) for adjusting imaging parameters with reference to a part of electronic images corresponding to the set detected by the detection means is further provided.

  Preferably, the specific object image corresponds to an image of a human face.

The image processing program according to the present invention, the processor (26) of the image processing apparatus comprises a capture means (30, 32) for capturing an electronic image output from the imaging means (16) (10), electrons captured by the capture means detecting a set of one or more specific object image from the image (S31 ~ S63, S89), a set of the cut-out area is detected by the detecting step having a size based on the size of the set which is detected by the detecting step Definition steps (S105 to S107) for defining the positions to be covered, and changing steps (S109 to S107) for changing the size and / or position of the clipping area when the clipping area defined by the defining step protrudes from the electronic image captured by the capturing means S 111), and cut, which is modified as necessary by the defined and changed step by defining step For executing a cut-out step (S1, S25, S27) of cutting out the electronic image captured by the capture means part of an electronic image belonging to the rear, an image processing program.

An image processing method according to the present invention is an image processing method that is executed by an image processing apparatus (10) including a capturing unit (30, 32) that captures an electronic image output from an imaging unit (16). A detection step (S31 to S63, S89) for detecting a set of one or more specific object images from the captured electronic image, and a cutout area having a size based on the size of the set detected by the detection step is detected by the detection step. A definition step (S105 to S107) for defining the position covering the set, and a change to change the size and / or position of the clipping area when the clipping area defined by the definition step protrudes from the electronic image captured by the capturing means Defined by step (S109 ~ S111) and definition step and changed as needed by change step A cutout step (S1, S25, S27) for cutting out a part of the electronic image belonging to the cutout area from the electronic image taken in by the taking-in means is provided.

According to the present invention, the position of the cutout area is adjusted so as to cover the set of specific object images, and the size of the cutout area is adjusted based on the scale of the set of specific object images. Thereby, the ratio of the background image to the set of specific object images is adaptively adjusted, and the reproducibility of the electronic image is improved.

  The above object, other objects, features and advantages of the present invention will become more apparent from the following detailed description of embodiments with reference to the drawings.

  Referring to FIG. 1, a digital camera 10 of this embodiment includes a focus lens 12 and an aperture unit 14 driven by drivers 18a and 18b, respectively. The optical image of the object scene that has passed through these members is irradiated onto the imaging surface of the imager 16 and subjected to photoelectric conversion. As a result, a charge representing the object scene image is generated.

  When the power is turned on, the CPU 26 instructs the driver 18c to repeat the exposure operation and the thinning readout operation in order to start the through image processing under the imaging task. In response to a vertical synchronization signal Vsync periodically generated from an SG (Signal Generator) (not shown), the driver 18c exposes the imaging surface and reads out part of the charges generated on the imaging surface in a raster scanning manner. From the imager 16, a low-resolution raw image signal based on the read charges is periodically output.

  The preprocessing circuit 20 performs processing such as CDS (Correlated Double Sampling), AGC (Automatic Gain Control), and A / D conversion on the raw image signal output from the imager 16, and outputs raw image data that is a digital signal. . The output raw image data is written into the raw image area 32 a of the SDRAM 32 through the memory control circuit 30.

  The post-processing circuit 34 reads the raw image data stored in the raw image area 32a through the memory control circuit 30, and performs processing such as white balance adjustment, color separation, and YUV conversion on the read raw image data. The YUV format image data thus generated is written into the YUV image area 32 b of the SDRAM 32 through the memory control circuit 30.

  The LCD driver 36 repeatedly reads the image data stored in the YUV image area 32b through the memory control circuit 30, and drives the LCD monitor 38 based on the read image data. As a result, a real-time moving image (through image) of the object scene is displayed on the monitor screen.

  Referring to FIG. 2, an evaluation area EVA is allocated at the center of the imaging surface. The evaluation area EVA is divided into 16 in each of the horizontal direction and the vertical direction, and 256 divided areas form the evaluation area EVA. In addition to the processing described above, the preprocessing circuit 20 executes simple RGB conversion processing that simply converts raw image data into RGB data.

  The AE / AWB evaluation circuit 22 integrates RGB data belonging to the evaluation area EVA among the RGB data generated by the preprocessing circuit 20 every time the vertical synchronization signal Vsync is generated. As a result, 256 integral values, that is, 256 AE / AWB evaluation values are output from the AE / AWB evaluation circuit 22 in response to the vertical synchronization signal Vsync.

  The AF evaluation circuit 24 extracts the high frequency component of the G data belonging to the same evaluation area EVA from the RGB data output from the preprocessing circuit 20, and the vertical synchronization signal Vsync is generated from the extracted high frequency component. Integrate every time. As a result, 256 integral values, that is, 256 AF evaluation values, are output from the AF evaluation circuit 24 in response to the vertical synchronization signal Vsync.

  The CPU 26 executes through image AE / AWB processing based on the output from the AE / AWB evaluation circuit 22 in parallel with the through image processing, and calculates an appropriate EV value and an appropriate white balance adjustment gain. The aperture amount and the exposure time that define the calculated appropriate EV value are set in the drivers 18b and 18c, respectively. The calculated appropriate white balance adjustment gain is set in the post-processing circuit 34. As a result, the brightness and white balance of the through image are appropriately adjusted.

  The CPU 26 also executes through-image AF processing based on the output from the AF evaluation circuit 24 under a continuous AF task in parallel with the through-image processing. The focus lens 12 is set to the focal point by the driver 18a when the output of the AF evaluation circuit 24 satisfies the AF activation condition. As a result, the focus of the through image is appropriately adjusted.

  When the shutter button 28s is half-pressed, the CPU 26 interrupts the continuous AF task and executes the recording AF process under the imaging task. The recording AF process is also executed based on the output of the AF evaluation circuit 24. As a result, the focus is adjusted strictly. Subsequently, the CPU 26 executes a recording AE process based on the output of the AE / AWB evaluation circuit 22 and calculates an optimum EV value. The aperture amount and the exposure time that define the calculated optimal EV value are set in the drivers 18b and 18c, respectively, as described above. As a result, the brightness of the through image is adjusted strictly.

  When the shutter button 28s is fully pressed, the CPU 26 instructs the driver 18c to execute an exposure operation and an all-pixel reading operation once for recording processing, and further activates the I / F 40. The driver 18c exposes the imaging surface in response to the vertical synchronization signal Vsync, and reads out all the charges generated thereby from the imaging surface in a raster scanning manner. From the imager 16, a one-frame raw image signal having a high resolution is output.

  The raw image signal output from the imager 16 is converted into raw image data by the preprocessing circuit 20, and the converted raw image data is written into the recorded image area 32 c of the SDRAM 32 by the memory control circuit 30. The CPU 26 calculates an optimum white balance adjustment gain based on the raw image data stored in the recorded image area 32c, and sets the calculated optimum white balance adjustment gain in the post-processing circuit 34.

  The post-processing circuit 34 reads the raw image data stored in the raw image area 32a through the memory control circuit 30 and converts the read raw image data into YUV format image data having an optimal white balance. The image data is written into the recorded image area 32c through the memory control circuit 30. The I / F 40 reads out the image data stored in the recording image area 32c in this way through the memory control circuit 30, and records the read image data in the recording medium 42 in a file format.

  Note that the through image processing is resumed when raw image data having high resolution is secured in the recorded image area 32c. The continuous AF task is also restarted at this point.

  The CPU 26 repeatedly searches for a human face image from low-resolution raw image data stored in the raw image area 32a of the SDRAM 32 under a face detection task executed in parallel with the through image processing. For such a face detection task, a dictionary DIC shown in FIG. 3, a plurality of face detection frames FD_1, FD_2, FD_3,..., And two tables TBL1 to TBL2 shown in FIG.

  According to FIG. 3, a plurality of face patterns FP_1, FP_2,... Are registered in the dictionary DIC. Further, according to FIG. 4, the face detection frames FD_1, FD_2, FD_3,... Have different shapes and / or sizes. Further, each of the tables TBL1 to TBL2 shown in FIG. 5 corresponds to a table for describing face frame information, and a column describing the position of the face image (the position of the face detection frame at the time when the face image is detected). And a column describing the size of the face image (the size of the face detection frame when the face image is detected).

  In the face detection task, first, the table TBL1 is designated as a current frame table that holds face frame information of the current frame. However, the specified table is updated between the tables TBL1 and TBL2 every frame, and the current frame table becomes the previous frame table in the next frame. When the specification of the current frame table is completed, the variable K is set to “1”, and the face detection frame FD_K is set to the upper left of the evaluation area EVA shown in FIG. 6, that is, the face detection start position.

  When the vertical synchronization signal Vsync is generated, among the raw image data of the current frame stored in the raw image area 32a of the SDRAM 32, partial image data belonging to the face detection frame FD_K is a plurality of data described in the dictionary DIC shown in FIG. Each of the face patterns FP_1, FP_2,. If it is determined that the focused partial image matches any face pattern, the current position and size of the face detection frame FD_K are described in the current frame table as face frame information.

  The face detection frame FD_K is moved by a predetermined amount in the raster direction in the manner shown in FIG. 6 and subjected to the above collation processing at a plurality of positions on the evaluation area EVA. Each time a person's face image is found, face frame information corresponding to the found face image (that is, the current position and size of the face detection frame FD_K) is described in the current frame table.

  When the face detection frame FD_K reaches the lower right of the evaluation area EVA, that is, the face detection end position, the variable K is updated, and the face detection frame FD_K corresponding to the updated value of the variable K is rearranged at the face detection start position. . As described above, the face detection frame FD_K moves in the raster direction on the evaluation area EVA, and the face frame information corresponding to the face image detected by the matching process is described in the current frame table. Such face recognition processing is repeatedly executed until the face detection frame FD_Kmax (Kmax: the number of the last face detection frame) reaches the face detection end position.

  When the face detection frame FD_Kmax reaches the face detection end position, the designation table is updated and the updated designation table is initialized. Further, the variable K is set to “1”. The face recognition process for the next frame is started in response to the generation of the vertical synchronization signal Vsync.

  In parallel with the imaging task described above, the CPU 40 determines the position and shape of the parameter adjustment area ADJ referred to for AE / AWB processing and AF processing, and the YUV image area of the SDRAM 32 for image reading processing by the LCD driver 36. The position and shape of the cut-out area CT assigned to 32b is defined under the area control task.

  In the area control task, the cut-out area CT is first initialized. The cutout area CT is set so as to cover the entire object scene image stored in the YUV image area 32b. The LCD driver 36 reads the image data belonging to the initial cut-out area CT from the YUV image area 32b, and displays the object scene image based on the read image data on the LCD monitor 38 as a through image.

  When the vertical synchronization signal Vsync is generated, the previous frame table in which the face frame information is determined is designated, and it is determined whether or not the face frame information is described in the previous frame table.

  If at least one face frame is described in the previous frame table, a face set frame GRP circumscribing the set of face frames is defined. Further, of the 256 divided areas forming the evaluation area EVA, a partial divided area that covers the area within the face collection frame GRP is defined as the parameter adjustment area ADJ. On the other hand, if one face image is not described in the previous frame table, the definition of the face collection frame GRP is canceled and the entire evaluation area EVA is defined as the parameter adjustment area ADJ.

  Therefore, when the rectangular frames KF1 and KF2 shown in FIG. 7 are described in the previous frame table, the face set frame GRP is defined so as to circumscribe the face frames KF1 and KF2. Further, a parameter adjustment area ADJ is defined so as to cover an area in the face collection frame GRP.

  The through image AE / AWB process and the recording AE / AWB process described above are AE / AWB evaluation values belonging to the parameter adjustment area ADJ among the 256 AE / AWB evaluation values output from the AE / AWB evaluation circuit 22. It is executed based on. The through-image AF process and the recording AF process are also executed based on the AF evaluation values belonging to the parameter adjustment area ADJ among the 256 AF evaluation values output from the AF evaluation circuit 24. This improves the accuracy of adjusting imaging parameters such as exposure amount and focus.

  When the shutter button 28s is operated (half-pressed), it is determined whether or not the face collection frame GRP is defined. If the face collection frame GRP is defined, the position and size of the cut-out area CT are adjusted as follows.

  First, the coefficients αtop, αunder, αleft, and αright are determined based on the size of the face collection frame GRP. Next, the coefficients βtop, βunder, βleft, and βright are determined based on the number of face frames described in the previous frame table.

  The coefficients αtop, αunder, αleft, and αright indicate different numerical values depending on the position and / or size of the face collection frame GRP. Also, the coefficients βtop, βunder, βleft, and βright indicate different numerical values depending on the number of face frames that form the face aggregate frame GRP. In particular, each of the coefficients βtop, βunder, βleft, and βright indicates “1.0” corresponding to the number of face frames = 1, and has a characteristic of approaching “0.0” as the number of face frames increases.

The coefficients αtop, αunder, αleft, αright and the coefficients βtop, βunder, βleft, and βright determined in this way are applied to Equation 1, thereby determining the position and size of the cut-out area CT.
[Equation 1]
D_top = D_fy * αtop * βtop
D_under = D_fy * αunder * βunder
D_left = D_fx * αleft * βleft
D_right = D_fx * αright * βright
However,
D_fx: Horizontal size of the face set frame GRP D_fy: Vertical size of the face set frame GRP D_top: Distance from the upper end of the face set frame GRP to the upper end of the cutout area CT D_under: From the lower end of the face set frame GRP to the lower end of the cutout area CT Distance D_left: distance from the left end of the face set frame GRP to the left end of the cutout area CT D_right: distance from the right end of the face set frame GRP to the right end of the cutout area CT

  The cut-out area CT defined by Equation 1 indicates the positional relationship and size relationship shown in FIG. 8A with the face collection frame GRP.

  However, the size of the cut-out area CT is corrected so as to conform to a predetermined aspect ratio (= 4: 3). Therefore, when the cutout area CT shown in FIG. 8A is obtained, the size of the cutout area CT is corrected as shown in FIG. 8B.

  When a cutout area CT that conforms to a predetermined aspect ratio is obtained, it is determined whether or not this cutout area CT protrudes from the YUV image area 32b (whether or not the cutout area CT matches an error condition). If the cutout area CT matches the error condition, the cutout area CT is redefined. That is, the size of the cutout area CT is reduced so that the cutout area CT fits in the YUV image area 32b.

  Therefore, when the cut-out area CT having a predetermined aspect ratio is arranged in the YUV image area 32b as shown in FIG. 9A, the size of the cut-out area CT is reduced as shown in FIG. 9B.

  When the shutter button 28s is in a half-pressed state, the LCD driver 36 reads out the image data belonging to the cutout area CT thus defined from the YUV image area 32b, and a part of the scene image based on the read image data Is fully displayed on the LCD monitor 38. Accordingly, the angle of view of the through image displayed on the LCD monitor 38 changes before and after the shutter button 28s is half-pressed.

  That is, when the shutter button 28s is half-pressed in a state where the object scene image shown in FIG. 10A is displayed on the LCD monitor 38, the face collection frame GRP and the cut-out area are processed as shown in FIG. CT is defined. On the LCD monitor 38, a part of the scene image belonging to the cutout area CT is enlarged and displayed.

  Further, when the shutter button 28s is half-pressed in a state where the object scene image shown in FIG. 11A is displayed on the LCD monitor 38, the face set frame GRP and the cutout area in the manner shown in FIG. 11B. CT is defined, and a part of the scene image belonging to the defined cutout area CT is enlarged and displayed on the LCD monitor 38.

  Further, when the shutter button 28s is half-pressed in a state where the object scene image shown in FIG. 12A is displayed on the LCD monitor 38, the face set frame GRP and the cut-out area as shown in FIG. CT is defined. As a result, a part of the scene image belonging to the cutout area CT is enlarged and displayed on the LCD monitor 38.

  When the shutter button 28s is fully pressed, the I / F 40 sets a recording cutout area (not shown) corresponding to the cutout area CT as the recording image area 32c, and some image data belonging to the recording cutout area Is read from the recording image area 32c, and the read image data is recorded on the recording medium.

  Thus, the position of the cutout area CT (and the recording cutout area) is adjusted to cover the set of face frames, and the size of the cutout area CT (and the cutout area for recording) is based on the size of the set of face frames. Adjusted. As a result, the ratio of the background image to the set of images of the human face is adaptively adjusted, and the reproducibility of the object scene image is improved.

  The CPU 40 has a plurality of tasks including an imaging task shown in FIGS. 13 to 14, a face detection task shown in FIGS. 15 to 17, a continuous AF task shown in FIG. 18, and an area control task shown in FIGS. Are executed in parallel. Control programs corresponding to these tasks are stored in the flash memory 44. The activation / stop of the face detection task and the continuous AF task is controlled by the imaging task.

  Referring to FIG. 13, through image processing is executed in step S1. As a result, a through image representing the scene is displayed on the LCD monitor 38. In step S3, a face detection task is activated, and in step S5, a continuous AF task is activated. In step S7, it is determined whether or not the shutter button 28s has been half-pressed, and as long as NO, the through image AE / AWB process in step S9 is repeated. The through image AE / AWB process is executed based on the AE / AWB evaluation values belonging to the parameter adjustment area ADJ, and thereby the brightness and white balance of the through image are appropriately adjusted.

  If “YES” in the step S7, the face detection task is stopped in a step S11, and the continuous AF task is stopped in a step S13. In step S15, a recording AF process is executed, and in step S17, a recording AE process is executed. The recording AF process is executed based on the AF evaluation value belonging to the parameter adjustment area ADJ, and the recording AE process is executed based on the AE / AWB evaluation value belonging to the parameter adjustment area ADJ. Thereby, the focus and brightness of the through image are strictly adjusted.

  In step S19, it is determined whether or not the shutter button 28s has been fully pressed. In step S21, it is determined whether or not the operation of the shutter button 28s has been released. If YES in step S19, the process proceeds to step S23, and if YES in step S21, the process returns to step S3. In step S23, a recording AWB process is executed, and in step S25, a recording process is executed. The recording AWB process is executed based on the AE / AWB evaluation values belonging to the parameter adjustment area ADJ. As a result, a high-resolution object scene image having an optimal white balance is recorded on the recording medium 42. In step S27, the through image processing is resumed, and then the process returns to step S3.

  Referring to FIG. 15, in step S31, it is determined whether or not the vertical synchronization signal Vsync is generated. If the determination result is updated from NO to YES, it is determined in step S33 whether or not the AF activation condition is satisfied. If “NO” here, the process directly returns to the step S31 while if “YES”, the through image AF process is executed in a step S35, and then, the process returns to the step S71.

  The process for determining whether or not the AF activation condition is satisfied is executed based on the AF evaluation value belonging to the parameter adjustment area ADJ, and the through image AF process is also executed based on the AF evaluation value belonging to the parameter adjustment area ADJ. . As a result, the focus of the through image is continuously adjusted.

  Referring to FIG. 16, in step S41, tables TBL1 and TBL2 are initialized, and in step S43, table TBL1 is designated as the current frame table. In step S45, the variable K is set to “1”, and in step S47, the face detection frame FD_K is arranged at the upper left face detection start position of the evaluation area EVA.

  The current frame table is updated between the tables TBL1 and TBL2 by the process of step S71 described later. Therefore, the current frame table becomes the previous frame table in the next frame.

  In step S49, it is determined whether or not the vertical synchronization signal Vsync is generated. If the determination result is updated from NO to YES, the variable L is set to “1” in step S51. In step S53, the partial image belonging to the face detection frame FD_K is collated with the face pattern FP_L registered in the dictionary DIC. In step S55, it is determined whether or not the partial image in the face detection frame FD_K matches the face pattern FP_L.

  If “NO” here, the variable L is incremented in a step S57, and it is determined in a step S59 whether or not the incremented variable L exceeds a constant Lmax (Lmax: total number of face patterns registered in the dictionary DIC). If L ≦ Lmax, the process returns to step S53, while if L> Lmax, the process proceeds to step S63.

  If “YES” in the step S55, the process proceeds to a step S61, and the current position and size of the face detection frame FD_K are described in the designation table as face frame information. When the process of step S61 is completed, the process proceeds to step S63.

  In step S63, it is determined whether or not the face detection frame FD_K has reached the lower right face detection end position of the evaluation area EVA. If “NO” here, the face detection frame FD_K is moved in the raster direction by a predetermined amount in a step S65, and thereafter, the process returns to the step S51. On the other hand, if “YES” in the step S63, the variable K is incremented in a step S67, and it is determined whether or not the incremented variable K exceeds “Kmax” in a step S69.

  If K ≦ Kmax, the process returns to step S47 while if K> Kmax, the process proceeds to step S71. In step S71, the specified table is updated and the updated specified table is initialized. When the process of step S71 is completed, the variable K is set to “1” in step S73, and then the process returns to step S47.

  Referring to FIG. 19, in step S81, the cutout area CT assigned to the YUV image area 32b of the SDRAM 32 is initialized. The cutout area CT is set so as to cover the entire object scene image stored in the YUV image area 32b.

  In step S83, it is determined whether or not the vertical synchronization signal Vsync has been generated. If the determination result is updated from NO to YES, the previous frame table is designated in step S85. In step S87, it is determined whether or not at least one face frame is described in the previous frame table. If YES, the process proceeds to step S85 through steps S89 to S91. If NO, the process proceeds to steps S93 to S95. Proceed to step S97.

  In step S89, the face collection frame GRP is defined so as to circumscribe the face frame described in the previous frame table. In step S91, of the 256 divided areas forming the evaluation area EVA, a partial divided area that covers the area within the face collection frame GRP is defined as the adjustment area ADJ. In step S93, the definition of the face collection frame GRP is canceled, and in step S95, the entire evaluation area EVA is defined as the adjustment area ADJ.

  In step S97, it is determined whether or not the shutter button 28s has been operated (half-pressed). If NO, the process returns to step S83, and if YES, the process proceeds to step S99. In step S99, it is determined whether or not the face collection frame GRP has been defined. If NO, the process proceeds directly to step S113, while if YES, the process proceeds to steps S113 through S111 to S111. In step S113, it is determined whether or not the operation of the shutter button 28s has been released. If the determination result is updated from NO to YES, the process returns to step S81.

  In step S101, the coefficients αtop, αunder, αleft, and αright are determined with reference to the size of the face collection frame GRP. In step S103, the coefficients βtop, βunder, βleft, and βright are determined with reference to the number of face frames detected from the object scene image of the previous frame.

  In step S105, the determined coefficients αtop, αunder, αleft, αright and the coefficients βtop, βunder, βleft, βright are applied to the above-described equation 1 to calculate the size and position of the cut-out area CT. In step S107, the size of the cut-out area CT calculated in step S105 is corrected so as to conform to a predetermined aspect ratio (= 4: 3).

  In step S109, it is determined whether or not the cutout area CT thus defined protrudes from the YUV image area 32b (whether or not the cutout area CT matches an error condition). If “NO” here, the process proceeds directly to the step S113, whereas if “YES”, the cut-out area CT is redefined in the step S111, and then the process proceeds to the step S113. As a result of the processing in step S111, the size of the cutout area CT is reduced so that the cutout area CT fits in the YUV image area 32b.

  As can be understood from the above description, the object scene image output from the imager 16 is taken into the SDRAM 32 by the memory control circuit 30. The CPU 26 detects a set of one or more face images from the captured scene image (S41 to S73, S89), and detects a cut-out area CT having a size based on the size of the detected set. The position is defined to cover the set (S105 to S111). The CPU 26 also cuts out a part of the scene image belonging to the cutout area CT from the scene image captured in the SDRAM 32 (S1, S25, S27).

  Thus, the position of the cutout area CT is adjusted so as to cover the set of face images, and the size of the cutout area CT is adjusted based on the scale of the set of face images. Thereby, the ratio of the background image to the set of face images is adaptively adjusted, and the reproducibility of the object scene image is improved.

  In this embodiment, the size of the cutout area CT is reduced when the cutout area CT matches the error condition. However, the position of the cutout area CT is changed instead of the size reduction or with the size reduction. You may make it do.

  Further, in this embodiment, a human face is assumed as the default object, but instead, an animal face such as a dog or a cat may be assumed.

  Further, in this embodiment, a so-called digital still camera that records still images is assumed, but the present invention can also be applied to a digital video camera that records moving images.

  In this embodiment, when the shutter button 28s is half-pressed, the face detection task is stopped (see steps S7 and S11 in FIG. 13), and the position and size of the cut-out area CT are determined (FIGS. 19 to 21). Steps S97 to S111). The position and size of the cutout area CT are fixed until the operation of the shutter button 28s is released.

  However, the face detection task may continue to be activated even after the shutter button 28s is half-pressed, and the position and size of the cut-out area CT may be changed according to the appearance, movement, and disappearance of the face image in the half-pressed state. In this case, preferably, the processing of step S11 shown in FIG. 13, step S97 shown in FIG. 19, and step S113 shown in FIG. 21 is omitted, and when NO is determined in step S99 of FIG. When the process of step S111 of 21 is completed, the process returns to step S83 shown in FIG.

  Furthermore, in this embodiment, it is defined as an error condition that the cut-out area CT protrudes from the YUV image area 32b, but in addition to this, the fact that the face that has been detected until now is no longer detected is also an error condition. You may make it prescribe | regulate.

It is a block diagram which shows the structure of one Example of this invention. It is an illustration figure which shows an example of the state which allocated the evaluation area to the imaging surface. It is an illustration figure which shows an example of the dictionary referred in FIG. 1 Example. It is an illustration figure which shows an example of the several face detection frame used for a face recognition process. It is an illustration figure which shows an example of the table referred by FIG. 1 Example. It is an illustration figure which shows a part of face detection operation | movement. It is an illustration figure which shows an example of the setting operation | movement of a parameter adjustment area. (A) is an illustration figure which shows a part of operation | movement which defines a cutout area, (B) is an illustration figure which shows another part of operation | movement which defines a cutout area. (A) is an illustration figure which shows the other part of operation | movement which defines a cut-out area, (B) is an illustration figure which shows another part of operation | movement which defines a cut-out area. (A) is an illustrative view showing an example of a scene, and (B) is an illustrative view showing an example of a face collection area and a cutout area assigned to the scene. (A) is an illustrative view showing another example of the object scene, and (B) is an illustrative view showing another example of the face collection area and the cutout frame allocated to the object scene. (A) is an illustrative view showing another example of the object scene, and (B) is an illustrative view showing another example of the face collection area and the cutout area assigned to the object scene. It is a flowchart which shows a part of operation | movement of CPU applied to the FIG. 1 Example. It is a flowchart which shows a part of other operation | movement of CPU applied to the FIG. 1 Example. It is a flowchart which shows a part of other operation | movement of CPU applied to the FIG. 1 Example. FIG. 12 is a flowchart showing yet another portion of behavior of the CPU applied to the embodiment in FIG. 1. It is a flowchart which shows a part of other operation | movement of CPU applied to the FIG. 1 Example. It is a flowchart which shows a part of other operation | movement of CPU applied to the FIG. 1 Example. FIG. 12 is a flowchart showing yet another portion of behavior of the CPU applied to the embodiment in FIG. 1. It is a flowchart which shows a part of other operation | movement of CPU applied to the FIG. 1 Example. It is a flowchart which shows a part of other operation | movement of CPU applied to the FIG. 1 Example.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 ... Digital camera 16 ... Image sensor 20 ... Pre-processing circuit 22 ... AE / AWB evaluation circuit 24 ... AF evaluation circuit 26 ... CPU
32 ... SDRAM
34 ... Post-processing circuit 44 ... Flash memory

Claims (8)

Capturing means for capturing an electronic image output from the imaging means;
Detecting means for detecting a set of one or more specific object images from the electronic image captured by the capturing means;
Defining means for defining a cutout area having a size based on the size of the set detected by the detecting means at a position covering the set detected by the detecting means;
Changing means for changing the size and / or position of the clipping area when the clipping area defined by the defining means protrudes from the electronic image captured by the capturing means; and defined by the defining means and required by the changing means comprising a clipping means for cutting out from the electronic image captured by the capturing means part of an electronic image belonging to the changed cutout area according to the image processing apparatus. The detecting means includes frame defining means for defining a frame in contact with a specific object image forming the set;
The image processing according to claim 1, wherein the definition unit includes a determination unit that determines a size and a position of the cut-out area based on the frame defined by the frame definition unit and the number of specific object images forming the set. apparatus. Further comprising display means for displaying the electronic image cut out by the cut-out means,
The imaging means repeatedly outputs the electronic image ,
Said fetching means fetches repeat the electronic image, the image processing apparatus according to claim 1 or 2 wherein. Further comprising initial setting means for initially setting the cutout area;
Said defining means performs to define processing in response to the default instruction after initialization by the initialization means, the image processing apparatus according to claim 3, wherein. The reference part of an electronic image corresponding to the set detected by the detection means further comprise adjusting means for adjusting imaging parameters, the image processing apparatus according to any one of claims 1 to 4. The specific object image corresponds to a face image portion of the person, the image processing apparatus according to any one of claims 1 to 5. In a processor of an image processing apparatus including a capturing unit that captures an electronic image output from an imaging unit,
A detection step of detecting a set of one or more specific object images from the electronic image captured by the capturing means;
A definition step for defining a cutout area having a size based on the scale of the set detected by the detection step at a position covering the set detected by the detection step;
A change step for changing the size and / or position of the cut-out area when the cut-out area defined by the definition step protrudes from the electronic image taken by the take-in means; and defined by the definition step and required by the change step an image processing program for executing the cutout step of part of an electronic image belonging to the changed cutout area according excised from electronic images captured by the capture means. An image processing method executed by an image processing apparatus including a capturing unit that captures an electronic image output from an imaging unit,
A detection step of detecting a set of one or more specific object images from the electronic image captured by the capturing means;
A definition step for defining a cutout area having a size based on the scale of the set detected by the detection step at a position covering the set detected by the detection step;
A change step for changing the size and / or position of the cut-out area when the cut-out area defined by the definition step protrudes from the electronic image taken by the take-in means; and defined by the definition step and required by the change step modified comprises cutout step of cutting out the electronic image captured by the capturing means part of an electronic image belonging to the cut-out area, the image processing method in accordance with.

Images