JP2012217106A - Image processing apparatus - Google Patents

Abstract

PROBLEM TO BE SOLVED: To enhance detection performance of the blurring of a plurality of images each capturing a common subject.SOLUTION: When a shutter button 28sh is pressed fully under a continuous shooting mode, a CPU 26 takes the image data of four consecutive frames, each capturing a common subject, into a save image area 32c of an SDRAM 32. The image data of each frame thus taken in is subjected to compression processing by means of a JPEG encoder 40. The CPU 26 detects the size value of the JPEG data of each frame as a parameter value correlated with the amount of high frequency image component, selects the maximum size value among four size values thus detected, and then normalizes the difference between the maximum size value thus selected and each of three remaining values.

Description

  The present invention relates to an image processing apparatus, and more particularly to an image processing apparatus that is applied to an electronic camera and detects the amount of blurring that appears in image data.

  An example of this type of device is disclosed in Patent Document 1. According to this background art, the magnitude of shake is detected for image data representing the subject captured by the image sensor, and the camera shake level is set based on the detected magnitude. The camera shake level is represented by a 20-level evaluation of levels 1 to 20 in the order from the smallest camera shake. As a camera shake detection method, a plurality of edges in image data are detected by image analysis, and the degree of camera shake is obtained from the detected feature amount of the edge, or the degree of camera shake is obtained from the output of the camera shake sensor. The method is adopted.

JP 2009-77237 A

  However, the former camera shake detection method causes a complicated software configuration, and the latter camera shake detection method causes a complicated hardware configuration. For this reason, in the background art, there is a limit to the improvement in the image data blur detection performance.

  SUMMARY OF THE INVENTION Therefore, a main object of the present invention is to provide an image processing apparatus that can improve the image blur detection performance.

  An image processing apparatus according to the present invention (10: reference numerals corresponding to the embodiments; the same applies hereinafter) includes acquisition means (S31 to S33, S37 to S39) for repeatedly acquiring images capturing a common subject, and the amount of high-frequency image components Detection means (S81) for detecting a parameter value correlated with each of the plurality of images acquired by the acquisition means, and selection means for selecting one of the plurality of parameter values detected by the detection means (S81). S83), and normalizing means (S91 to S99, S103, S109 to S111) for normalizing the difference between the parameter value selected by the selecting means and each of one or more parameter values remaining after selection of the selecting means ).

  Preferably, the parameter value selected by the selection means corresponds to an image having a maximum amount of high-frequency image components.

  Preferably, the normalization means creates a reference blur image having a reference blur amount based on an image corresponding to the parameter value selected by the selection means (S91), the reference blur image created by the creation means Reference parameter value detection means (S93 to S95) for detecting a parameter value correlated with a high-frequency image component as a reference parameter value, and normalization that performs normalization processing with reference to the reference parameter value detected by the reference parameter value detection means Processing means (S103).

  Preferably, an encoding unit (S35) for individually encoding a plurality of images acquired by the acquiring unit is further provided, and the parameter value detected by the detecting unit is set to the size of the image encoded by the encoding unit. Equivalent to.

  Preferably, extraction means (S105, S49 to S73) for extracting a part of the plurality of images acquired by the acquisition means based on the normalized difference calculated by the normalization means is further provided.

  More preferably, the extracting means searches for a specific object image from the image acquired by the acquiring means (S55), a predetermined calculation based on the search result of the searching means and the normalized difference calculated by the normalizing means. Calculation means (S57 to S65) for executing processing, and extraction processing means (S71 to S73) for executing extraction processing with reference to the calculation result of the calculation means are included.

  The image processing program according to the present invention includes an acquisition step (S31 to S33, S37 to S39) for repeatedly acquiring an image capturing a common subject in the processor (26) of the image processing apparatus (10), and the amount of high-frequency image components. A detection step (S81) for detecting correlated parameter values corresponding to each of the plurality of images acquired by the acquisition step, and a selection step (S83) for selecting any one of the plurality of parameter values detected by the detection step ), And a normalization step (S91 to S99, S103, S109 to S111) for normalizing the difference between the parameter value selected by the selection step and each of one or more parameter values remaining after selection of the selection step Is an image processing program for executing.

  The image processing method according to the present invention is an image processing method executed by the image processing apparatus (10), and repeatedly obtains images capturing a common subject (S31 to S33, S37 to S39), a high-frequency image A detection step (S81) for detecting a parameter value correlated with the amount of component corresponding to each of the plurality of images acquired by the acquisition step, and selecting any one of the plurality of parameter values detected by the detection step A selection step (S83), and a normalization step (S91 to S99, S103, S103, S103, S103, S103, S103, S103, S103, S103, S103, S103, S103, and S103) S109 to S111).

  An external control program according to the present invention is an external control program supplied to an image processing apparatus including a processor (26) that executes processing according to an internal control program stored in a memory (46), and captures a common subject. Acquisition step for repeatedly acquiring images (S31 to S33, S37 to S39), detection step for detecting a parameter value correlated with the amount of high-frequency image components corresponding to each of the plurality of images acquired by the acquisition step (S81) A selection step (S83) for selecting any one of a plurality of parameter values detected by the detection step, and a parameter value selected by the selection step and one or more parameter values remaining after selection of the selection step Normalization steps (S91 to S99, S103, S109 to S111) for normalizing the difference from each of them are executed by the processor in cooperation with the internal control program Of the eye, which is an external control program.

  An image processing apparatus (10) according to the present invention executes a process according to a fetching means (48) for fetching an external control program, and an external control program fetched by the fetching means and an internal control program stored in a memory (46). An image processing apparatus (10) comprising a processor (26), wherein an external control program repeatedly acquires images capturing a common subject (S31 to S33, S37 to S39), and determines the amount of high-frequency image components A detection step (S81) for detecting correlated parameter values corresponding to each of the plurality of images acquired by the acquisition step, and a selection step (S83) for selecting any one of the plurality of parameter values detected by the detection step ), And the difference between the parameter value selected by the selection step and each of the one or more parameter values remaining after selection of the selection step Normalization step of normalizing the (S91 ~ S99, S103, S109 ~ S111) corresponds to the internal control program in cooperation with a program to be executed.

  The parameter value that correlates with the amount of the high-frequency image component varies due to the attribute of the subject, and also varies due to blurring of the imaging surface at the time of image acquisition. However, fluctuations caused by subject attributes are eliminated by acquiring a plurality of images capturing a common subject and normalizing the above-described parameter values corresponding to the images.

  Based on this, in the present invention, any one of a plurality of parameter values respectively corresponding to a plurality of images capturing a common subject is selected, and the selected parameter value and one or two or more remaining after the selection are selected. The difference with each of the parameter values is normalized. This improves the blur detection performance of a plurality of images capturing a common subject.

  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.

It is a block diagram which shows the basic composition of one Example of this invention. 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 allocation state of the evaluation area in an imaging surface. FIG. 3 is an illustrative view showing one example of a configuration of a register applied to the embodiment in FIG. 2; It is an illustration figure which shows an example of the scene caught on an imaging surface. (A) is an illustrative view showing an example of image data of a first frame, and (B) is an illustrative view showing an example of image data of a second frame. (A) is an illustration figure which shows an example of the image data of a 3rd frame, (B) is an illustration figure which shows an example of the image data of a 4th frame. It is an illustration figure which shows an example of the blurring image data produced based on reference | standard image data. It is an illustration figure which shows an example of a structure of the face dictionary referred in a face detection process. FIG. 10 is an illustrative view showing one example of a configuration of another register applied to the embodiment in FIG. 2; It is a flowchart which shows a part of operation | movement of CPU applied to the FIG. 2 Example. It is a flowchart which shows a part of other operation | movement of CPU applied to the FIG. 2 Example. FIG. 11 is a flowchart showing still another portion of behavior of the CPU applied to the embodiment in FIG. 2; FIG. 10 is a flowchart showing yet another portion of behavior of the CPU applied to the embodiment in FIG. 2; It is a flowchart which shows a part of other operation | movement of CPU applied to the FIG. 2 Example. FIG. 11 is a flowchart showing still another portion of behavior of the CPU applied to the embodiment in FIG. 2; It is a block diagram which shows the structure of the other Example of this invention.

Embodiments of the present invention will be described below with reference to the drawings.
[Basic configuration]

  Referring to FIG. 1, the image processing apparatus of this embodiment is basically configured as follows. The acquisition unit 1 repeatedly acquires images that capture a common subject. The detection unit 2 detects a parameter value correlated with the amount of the high frequency image component corresponding to each of the plurality of images acquired by the acquisition unit 1. The selection unit 3 selects any one of a plurality of parameter values detected by the detection unit 2. The normalizing unit 4 normalizes the difference between the parameter value selected by the selecting unit 3 and each of one or more parameter values remaining after selection by the selecting unit 3.

  The parameter value that correlates with the amount of the high-frequency image component varies due to the attribute of the subject, and also varies due to blurring of the imaging surface at the time of image acquisition. However, fluctuations caused by subject attributes are eliminated by acquiring a plurality of images capturing a common subject and normalizing the above-described parameter values corresponding to the images.

Based on this, in this apparatus, one of a plurality of parameter values respectively corresponding to a plurality of images capturing a common subject is selected, and the selected parameter value and one or two or more remaining after the selection are selected. The difference with each of the parameter values is normalized. This improves the blur detection performance of a plurality of images capturing a common subject.
[Example]

  Referring to FIG. 2, the 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 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 scene captured on the imaging surface is generated.

  When the power is turned on, the CPU 26 instructs the driver 18c to repeat the exposure operation and the charge readout operation under the imaging task in order to execute the moving image capturing process. 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 the charges generated on the imaging surface in a raster scanning manner. From the imager 16, raw image data based on the read charges is periodically output.

  The preprocessing circuit 20 performs processing such as digital clamping, pixel defect correction, and gain control on the raw image data output from the imager 16. The raw image data subjected to these processes 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 color separation processing, white balance adjustment processing, and YUV conversion processing 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 by 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 moving image representing a scene captured on the imaging surface is displayed on the monitor screen in real time.

  Referring to FIG. 3, 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 above-described processing, the preprocessing circuit 20 illustrated in FIG. 2 executes simple RGB conversion processing that simply converts raw image data into RGB data (RGB: red, green & blue).

  The AE 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 evaluation values, are output from the AE evaluation circuit 22 in response to the vertical synchronization signal Vsync. The AF evaluation circuit 24 integrates the high-frequency components of the 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 AF evaluation values, are output from the AF evaluation circuit 24 in response to the vertical synchronization signal Vsync.

  When the shutter button 28sh provided in the key input device 28 is in a non-operating state, the CPU 26 executes a simple AE process based on the 256 AE evaluation values output from the AE evaluation circuit 22, and sets an appropriate EV value. calculate. The aperture amount and exposure time that define the calculated appropriate EV value are set in the drivers 18b and 18c, whereby the brightness of the through image is roughly adjusted.

  When the shutter button 28sh is half-pressed, the CPU 26 executes a strict AE process with reference to the AE evaluation value, and calculates an optimum EV value. The aperture amount and the exposure time that define the calculated optimum EV value are also set in the drivers 18b and 18c, whereby the brightness of the through image is strictly adjusted. The CPU 26 also executes AF processing based on the 256 AF evaluation values output from the AF evaluation circuit 24. The focus lens 12 is moved in the optical axis direction by the driver 18a in order to search for a focal point, and is arranged at the focal point found by this. As a result, the sharpness of the through image is improved.

  The imaging mode is switched between the single shooting mode and the continuous shooting mode by the mode switch 28md. In the continuous shooting mode, the best shot select function can be turned on / off.

  When the shutter button 28sh is fully pressed while the single shooting mode is selected, the CPU 26 executes the still image capturing process only once. As a result, one frame of image data representing the scene at the time when the shutter button 28sh is fully pressed is saved in the save image area 32c. When the still image capturing process is completed, the CPU 26 instructs the JPEG encoder 40 to execute the compression process, and instructs the memory I / F 42 to execute the recording process.

  The JPEG encoder 40 reads the image data saved in the saved image area 32 c through the memory control circuit 30, compresses the read image data by the JPEG method, and compresses the compressed image data, that is, JPEG data through the memory control circuit 30 to the SDRAM 32. Are written in the compressed image area 32d. The memory I / F 42 reads the JPEG data stored in the compressed image area 32d through the memory control circuit 30, and records the read JPEG data on the recording medium 44 in a file format.

  When the shutter button 28sh is fully pressed in the state where the continuous shooting mode is selected, the CPU 26 executes a total of four still image capturing processes every time the vertical synchronization signal Vsync is generated, and four more JPEG compression processes. Is executed to the JPEG encoder 40.

  As a result of the four still image capturing processes, four consecutive frames of image data are saved in the save image area 32c. The JPEG encoder 40 reads the saved 4-frame image data through the memory control circuit 30, compresses the read-out image data of each frame by the JPEG method, and compresses the JPEG data through the memory control circuit 30 to a compressed image area 32d. Write to. The compressed image area 32d stores JPEG data of a total of 4 frames.

  When the best shot select function is in the OFF state, the CPU 26 instructs the memory I / F 42 to execute recording processing for all the four frames of JPEG data stored in the compressed image area 32d. On the other hand, if the best shot select function is in the ON state, the CPU 26 selects one frame from the four frames stored in the compressed image area 32d, and executes the recording process for the JPEG data of the selected frame. To the memory I / F 42. The memory I / F 42 reads the JPEG data of the selected frame from the compressed image area 32d through the memory control circuit 30, and records the read JPEG data on the recording medium 44 in a file format.

  When the best shot select function is on, the CPU 26 executes the following processing.

  First, a blur determination process is executed to determine a camera shake on the imaging surface immediately after the shutter button 28sh is fully pressed. In the blur determination process, the maximum size value is specified from four size values respectively corresponding to the four frames of JPEG data stored in the compressed image area 32d, and uncompressed corresponding to the JPEG data of the specified size value. Image data is designated as reference image data. The frame number of the designated reference image data is set to the variable R, the flag FLG_R_blur is set to “0”, and the stability of the reference image data is set to “100”. The stability “100” is registered in the register RGST1 shown in FIG. 4 corresponding to the R-th frame.

  As shown in FIG. 5, when the shutter button 28sh is fully pressed in a state where the scene in which the person HM1, the building BL1 and the mountain MT1 appear is captured, and the camera shake occurs on the imaging surface due to this, the first frame 6A is created in the manner shown in FIG. 6A, the second frame image data is created in the manner shown in FIG. 6B, and the third frame image data is created in the manner shown in FIG. The image data of the fourth frame is created in the manner shown in FIG. 7B.

  Among the four frames of image data obtained in this way, the image data of the third frame is the image data with the smallest camera shake, and the size value of the JPEG data of the third frame is the maximum. Therefore, in this example, the image data of the third frame is designated as the reference image data, and the variable R is set to “3”. Further, the flag FLG — 3_blur is set to “0”, and the stability of the image data of the third frame is set to “100”.

  Subsequently, blurred image data having a reference blur degree (= the maximum allowable blur degree) is created based on the reference image data. Specifically, the reference image data is shifted or rotated by a predetermined amount in the horizontal direction, the vertical direction, and the direction around the optical axis, and a plurality of frames of reference image data defined at different positions or postures are synthesized with each other. The blurred image data is created in the manner shown in FIG. 8 based on the image data shown in FIG. The created blurred image data is stored in the saved image area 32 c of the SDRAM 32.

  The CPU 26 instructs the JPEG encoder 40 to execute compression processing on the blur image data created in this way. The JPEG encoder 40 reads the blurred image data from the saved image area 32 c through the memory control circuit 30, compresses the read blurred image data by the JPEG method, and compresses the JPEG data based on the blurred image data through the memory control circuit 30. Write to the image area 32d. The size value of the created JPEG data is set to the reference value BLURsz.

  Subsequently, the variable K is sequentially set to “1” to “4” (except for the same value as the variable R), and the size value of the JPEG data of the Kth frame is compared with the reference value BLURsz. If the size value of interest is greater than or equal to the reference value BLURsz, the flag FLG_K_blur is set to “0”. On the other hand, if the size value of interest falls below the reference value BLURsz, the flag FLG_K_blur is set to “1”. That is, the flag FLG_K_blur is set to “1” when the blurring degree of the image data of the Kth frame falls outside the allowable range, and is set to “0” when the blurring degree of the image data of the Kth frame falls within the allowable range. .

  In the example of FIGS. 6A to 7B, the size of the JPEG data of the first frame is less than the reference value BLURsz, and the sizes of the JPEG data of the second frame and the fourth frame are greater than or equal to the reference value BLURsz. Therefore, the flag FLG_1_blue is set to “1”, and the flags FLG_2_blue and FLG_4_blue are set to “0”.

Further, when the size value of interest is equal to or greater than the reference value BLURsz, the blurring degree of the image data of the Kth frame is calculated according to Equation 1, and the stability of the image data of the Kth frame is calculated according to Equation 2.
[Equation 1]
Blur degree = (REFsz−Ksz) / (REFsz−BLURsz)
Ksz: JPEG data size value of the Kth frame REFsz: JPEG data size value based on the reference image data [Expression 2]
Stability = 1-blurness

  According to Equation 1, the difference between the size value of the JPEG data of the Kth frame and the size value of the JPEG data based on the reference image data is the difference between the size value of the JPEG data based on the reference image data and the JPEG data based on the blurred image data. Divide by the difference from the size value. Thereby, the difference between the size value of the JPEG data of the Kth frame and the size value of the JPEG data based on the reference image data is normalized. According to Equation 2, the degree of blur calculated by Equation 1 is subtracted from “1”. The stability calculated according to Equation 2 is registered in the register RGST1 corresponding to the Kth frame.

  When the blur determination process is completed, the variable K is set to “1” to “4” again, and the state of the flag FLG_K_blue is determined.

  If FLG_K_blue = 1, the overall evaluation score is set to “0” because the blurring degree of the image data of the Kth frame is out of the allowable range. The overall evaluation score “0” is described in the register RGST1 corresponding to the Kth frame. In the above example, the flag FLG_1_blue is set to “1” corresponding to the image data of the first frame shown in FIG. Therefore, the overall evaluation score of the first frame is set to “0”.

  On the other hand, if FLG_K_blue = 0, the face detection process is performed on the image data of the Kth frame saved in the saved image area 32c. In the face detection process, a face image matching one of the three dictionary images stored in the manner shown in FIG. 9 in the face dictionary DC_F in the flash memory 46 is searched from the image data of the Kth frame. When one or more face images are found by the search process, the size and position of the face frame surrounding the found face image are registered in the register RGST2 shown in FIG. The number of face images found is also registered in the register RGST2.

  Accordingly, the size and position of the face frame indicated by “FK” in FIG. 6B is registered for the image data of the second frame, and the image data of the third frame is indicated by “FK” in FIG. 7A. The size and position of the face frame are registered, and the size and position of the face frame indicated by “FK” in FIG. 7B are registered for the image data of the fourth frame. In any of the second to fourth frames, the number of registered face images indicates “1”.

  If the number of face images registered in the register RGST2 is “1” or more, the smile level is calculated for images belonging to the face frame registered in the register RGST1. Further, eyes are detected on an image belonging to the face frame registered in the register RGST1, and the opening degree of the detected eyes is calculated. If the number of face images found is “2” or more, the calculated smile level is equivalent to the average smile level of each face image, and the calculated opening degree is the opening degree of the eye appearing in each face image. It corresponds to the average value of degrees. Both the calculated smile degree and eye opening indicate values belonging to the range of “0” to “1”. On the other hand, if the number of face images registered in the register RGST2 is “0”, the smile level and the eye opening are set to “0”. The smile level and the eye opening thus obtained are registered in the register RGST1 shown in FIG. 4 corresponding to the Kth frame.

Subsequently, the stability, smile degree, and eye opening degree registered in the register RGST1 corresponding to the Kth frame are applied to Equation 3, thereby calculating a comprehensive evaluation score.
[Equation 3]
Overall score = Stability * Wstbl + Smile * Wsml + Eye opening * Weey
However, Wstbl>Wsml>Weye> 0

  According to Equation 3, the stability is multiplied by the weighting amount Wstbl, the smile degree is multiplied by the weighting amount Wsml, the eye opening is multiplied by the weighting amount Weye, and these multiplication values are added to each other. The total evaluation score calculated in this way is also described in the register RGST1 shown in FIG. 4 corresponding to the Kth frame.

  When four total evaluation points are registered in the register RGST1, the highest total evaluation point is detected from these total evaluation points, and a frame corresponding to the detected total evaluation point is selected. In the above example, the total evaluation score corresponding to the fourth frame is maximized, and the fourth frame is selected. The CPU 26 instructs the memory I / F 42 to record JPEG data corresponding to the selected frame. The memory I / F 42 reads the JPEG data of the designated frame from the compressed image area 32d through the memory control circuit 30, and records the read JPEG data on the recording medium 44 in a file format.

  The CPU 26 executes a plurality of tasks including the imaging tasks shown in FIGS. 11 to 16 in parallel under the control of the multitask OS. Note that control programs corresponding to these tasks are stored in the flash memory 46.

  Referring to FIG. 11, in step S1, a moving image capturing process is executed. As a result, a through image representing a scene is displayed on the LCD monitor 38. In step S3, it is determined whether or not the shutter button 28sh is half-pressed. As long as the determination result is NO, the simple AE process is repeated in step S5. As a result, the brightness of the through image is roughly adjusted. When the determination result in step S3 is updated from NO to YES, the strict AE process is executed in step S7, and the AF process is executed in step S9. The brightness of the through image is strictly adjusted by the strict AE process, and the sharpness of the through image is improved by the AF process.

  In step S11, it is determined whether or not the shutter button 28sh has been fully pressed. In step S13, it is determined whether or not the operation of the shutter button 28sh has been released. If “YES” in the step S113, the process returns to the step S3 as it is, and if “YES” in the step S11, the process returns to the step S3 through steps S15 to S23.

  In step S15, it is determined whether the current imaging mode is the continuous shooting mode or the single shooting mode. If the current imaging mode is the single shooting mode, a still image capturing process is executed in step S17. In step S19, the JPEG encoder 40 is instructed to execute compression processing, and in step S21, the memory I / F 42 is instructed to execute recording processing. On the other hand, if the current imaging mode is the continuous shooting mode, continuous shooting & recording processing is executed in step S23. When the process of step S21 or S23 is completed, the process returns to step S3.

  As a result of the still image capturing process in step S17, one frame of image data representing the scene at the time when the shutter button 28sh is fully pressed is saved from the YUV image area 32b to the save image area 32c. Further, as a result of the processing in step S19, JPEG data based on the saved image data is created by the JPEG encoder 40. Further, as a result of the processing in step S21, the created JPEG data is recorded on the recording medium 44 in a file format.

  The continuous shooting and recording process in step S23 is executed according to the subroutine shown in FIGS.

  First, in step S31, a variable K corresponding to the frame number is set to “1”. In step S33, the still image capturing process similar to that in step S17 described above is executed after the vertical synchronization signal Vsync is generated. In step S35, the JPEG encoder 40 is instructed to execute compression processing, and in step S37, the variable K is incremented. In step S39, it is determined whether or not the incremented variable K exceeds “4”. If the determination result is NO, the process returns to step S33, while if the determination result is YES, the process proceeds to step S41 and subsequent steps.

  Therefore, the still image capturing process in step S33 and the JPEG compression process in step S35 are executed a total of four times in response to the vertical synchronization signal Vsync, and continuous four frames of image data are saved in the save image area 32c. Based on this, four frames of JPEG data are stored in the compressed image area 32d.

  In step S41, it is determined whether or not the best shot select function is on. If the determination result is NO, the process proceeds to step S43, and the memory I / F 42 is commanded to execute the recording process for the 4-frame image data stored in the compressed image area 32d. The memory I / F 42 reads four frames of JPEG data from the compressed image area 32d through the memory control circuit 30, and records the read JPEG data on the recording medium 42 in a file format. When the recording process is completed, the routine returns to the upper layer routine.

  If the decision result in the step S41 is YES, the flags FLG_1_blur to FLG_4_blur are set to “0” in a step S45, and a blur decision process is executed in a step S47. The flag FLG_K_blur (K: 1 to 4) maintains “0” when the blurring degree of the image data of the Kth frame falls within the allowable range, and “1” when the blurring degree of the image data of the Kth frame falls outside the allowable range. Is updated. For image data with FLG_K_blur = 0, the stability of the image is calculated, and the calculated stability is registered in the register RGST1 corresponding to the Kth frame.

  In step S49, the variable K is set to “1”, and in step S51, it is determined whether or not the flag FLG_K_blue indicates “1”. If the determination result is YES, the process proceeds to step S53, and the comprehensive evaluation point “0” is described in the register RGST1 corresponding to the Kth frame. When the description is completed, the process proceeds to step S67.

  On the other hand, if the determination result of step S51 is NO, the face detection process for the image data of the Kth frame is executed in step S55. As a result, a face image that matches one of the dictionary images stored in the face dictionary DC_F is searched from the image data of the Kth frame. When one or more face images are found by the search process, the size and position of the face frame surrounding the found face image are registered in the register RGST2. The number of face images found is also registered in the register RGST2.

  In step S57, it is determined whether or not the number of face images registered in the register RGST2 is “0”. If the determination result is NO, the process proceeds from step S59 to S61 through step S65, and if the determination result is YES, the process proceeds from step S63 to step S65.

  In step S59, the smile level is calculated with reference to images belonging to the face frame registered in the register RGST1. In step S61, eyes are detected on an image belonging to the face frame registered in the register RGST1, and the opening degree of the detected eyes is calculated. If the number of face images found is “2” or more, the smile level calculated in step S59 corresponds to the average smile level of each face image, and the degree of opening calculated in step S61 is the face image. It corresponds to the average value of the eye opening. In step S63, the smile level and the eye opening are set to “0”. The smile level and the eye opening obtained in this way are registered in the register RGST1 corresponding to the Kth frame.

  In step S65, the overall evaluation score is calculated by applying the stability, smile degree, and eye opening registered in the register RGST1 corresponding to the Kth frame to Equation 3. The calculated overall evaluation score is also described in the register RGST1 corresponding to the Kth frame. When the description is completed, the process proceeds to step S67.

  In step S67, it is determined whether or not the variable K has reached “4”. If the determination result is NO, the variable K is incremented in step S69 and the process returns to step S51. Proceed to S71. In step S71, the highest total evaluation score is detected from the four total evaluation scores registered in the register RGST1, and a frame corresponding to the detected total evaluation score is selected.

  In step S73, the memory I / F 42 is commanded to record JPEG data corresponding to the selected frame. The memory I / F 42 reads the JPEG data of the designated frame from the compressed image area 32d through the memory control circuit 30, and records the read JPEG data on the recording medium 44 in a file format. When the process of step S73 is completed, the process returns to the upper layer routine.

  The blur determination process in step S47 shown in FIG. 12 is executed according to a subroutine shown in FIGS. First, in step S81, the size value of JPEG data of each frame is detected. In step S83, the maximum size value is specified from the four detected size values, and uncompressed image data corresponding to the JPEG data having the specified size value is designated as the reference image data.

  In step S85, the frame number of the designated reference image data is set to the variable R, in step S87, the flag FLG_R_blur is set to “0”, and in step S89, the stability of the reference image data is set to “100”. The degree of stability “100” is registered in the register RGST1 corresponding to the Rth frame. In step S91, blur image data having a reference blur degree (= maximum allowable blur degree) is created based on the reference image data. The created blurred image data is stored in the saved image area 32 c of the SDRAM 32.

  In step S93, the JPEG encoder 40 is instructed to execute compression processing on the created blurred image data. The JPEG encoder 40 reads the blurred image data from the saved image area 32 c through the memory control circuit 30, compresses the read blurred image data by the JPEG method, and compresses the compressed image data, that is, JPEG data through the memory control circuit 30. Write to area 32d. In step S95, the size value of the JPEG data created in step S93 is set as the reference value BLURsz.

  In step S97, the variable K is set to “1”, and in step S99, it is determined whether or not the variable K matches the variable R. If a determination result is YES, it will progress to step S109 as it is, and if a determination result is NO, it will progress to step S109 through the process of steps S101-S107.

  In step S101, it is determined whether or not the size value of the JPEG data of the Kth frame is below the reference value BLURsz. If the determination result is NO, the flag FLG_K_blur is set to “1” in a step S107. If the determination result is YES, the blurring degree of the image data of the Kth frame is calculated in step S103, and the stability of the image data of the Kth frame is calculated in step S105. The degree of blur is calculated according to Equation 1, and the stability is calculated according to Equation 2. The stability calculated according to Equation 2 is registered in the register RGST1 corresponding to the Kth frame.

  In step S109, it is determined whether or not the variable K has reached “4”. If the determination result is NO, the variable K is incremented in step S111 and then the process returns to step S99. Return to the hierarchy routine.

  As can be seen from the above description, when the shutter button 28sh is fully pressed under the continuous shooting mode, the CPU 26 captures continuous four frames of image data capturing a common subject into the saved image area 32c of the SDRAM 32 (S31). ~ S33, S37 ~ S39). The captured image data of each frame is subjected to compression processing by the JPEG encoder 40. The CPU 26 detects the size value of the JPEG data of each frame as a parameter value correlating with the amount of the high frequency image component (S81), and selects the maximum size value from the four size values thus detected (S83), Then, the difference between the selected maximum size value and each of the remaining three size values is normalized (S91 to S99, S103, S109 to S111).

  The size value of the JPEG data varies due to the attribute of the subject, and also varies due to blurring of the imaging surface at the time of image acquisition. However, fluctuations caused by subject attributes are eliminated by acquiring a plurality of images capturing a common subject and normalizing the size value of JPEG data corresponding to each image.

  Based on this, in this embodiment, the maximum size value is selected from four size values respectively corresponding to four frames of JPEG data representing a common subject, and the selected maximum size value and the remaining three sizes are selected. The difference with each of the values is normalized. This improves the blur detection performance of a plurality of images capturing a common subject.

  In this embodiment, JPEG data of the frame having the highest overall evaluation score is recorded (see steps S71 to S73 in FIG. 14). However, as long as the overall evaluation score exceeds the reference score, two or more frames are recorded. The JPEG data may be recorded.

  In this embodiment, the multitask OS and control programs corresponding to a plurality of tasks executed thereby are stored in the flash memory 46 in advance. However, as shown in FIG. 17, a communication I / F 48 is provided in the digital camera 10, and some control programs are prepared as internal control programs in the flash memory 46 from the beginning, while some other control programs are prepared as external control programs. May be acquired from an external server. In this case, the above-described operation is realized by cooperation of the internal control program and the external control program.

  In this embodiment, the process executed by the CPU 26 is divided into a plurality of tasks as described above. However, each task may be further divided into a plurality of small tasks, and a part of the divided plurality of small tasks may be integrated with other tasks. Further, when each task is divided into a plurality of small tasks, all or part of the tasks may be acquired from an external server.

DESCRIPTION OF SYMBOLS 10 ... Digital camera 16 ... Imager 22 ... AE evaluation circuit 24 ... AF evaluation circuit 26 ... CPU
32 ... SDRAM
40: JPEG encoder 46: Flash memory

Claims (10)

Acquisition means for repeatedly acquiring images capturing a common subject;
Detecting means for detecting a parameter value correlated with the amount of the high-frequency image component corresponding to each of the plurality of images acquired by the acquiring means;
A selection means for selecting any one of a plurality of parameter values detected by the detection means; and a parameter value selected by the selection means and one or more parameter values remaining after selection by the selection means An image processing apparatus comprising normalizing means for normalizing a difference from each.   The image processing apparatus according to claim 1, wherein the parameter value selected by the selection unit corresponds to an image having a maximum amount of the high-frequency image component.   The normalizing means creates a reference blurred image having a reference blur amount based on an image corresponding to the parameter value selected by the selecting means, and a high-frequency image component of the reference blurred image created by the creating means A reference parameter value detecting means for detecting a parameter value correlated with the reference parameter value, and a normalization processing means for executing a normalization process with reference to the reference parameter value detected by the reference parameter value detecting means. Item 3. The image processing apparatus according to Item 1 or 2. An encoding unit that individually encodes the plurality of images acquired by the acquiring unit;
4. The image processing apparatus according to claim 1, wherein the parameter value detected by the detection unit corresponds to a size of an image encoded by the encoding unit.   5. The image processing apparatus according to claim 1, further comprising an extracting unit that extracts a part of the plurality of images acquired by the acquiring unit based on a normalized difference calculated by the normalizing unit. .   The extraction unit searches for a specific object image from the image acquired by the acquisition unit, and executes a predetermined calculation process based on a search result of the search unit and a normalized difference calculated by the normalization unit. The image processing apparatus according to claim 5, further comprising: an arithmetic processing unit that performs an extraction process with reference to a calculation result of the arithmetic unit. In the processor of the image processing device,
An acquisition step of repeatedly acquiring images capturing a common subject;
A detecting step for detecting a parameter value correlated with the amount of the high-frequency image component corresponding to each of the plurality of images acquired by the acquiring step;
A selection step of selecting any one of a plurality of parameter values detected by the detection step, and a parameter value selected by the selection step and one or more parameter values remaining after selection of the selection step An image processing program for executing a normalization step for normalizing a difference from each. An image processing method executed by an image processing apparatus,
An acquisition step of repeatedly acquiring images capturing a common subject;
A detecting step for detecting a parameter value correlated with the amount of the high-frequency image component corresponding to each of the plurality of images acquired by the acquiring step;
A selection step of selecting any one of a plurality of parameter values detected by the detection step, and a parameter value selected by the selection step and one or more parameter values remaining after selection of the selection step An image processing method comprising a normalization step of normalizing a difference from each. An external control program supplied to an image processing apparatus including a processor that executes processing according to an internal control program stored in a memory,
An acquisition step of repeatedly acquiring images capturing a common subject;
A detecting step for detecting a parameter value correlated with the amount of the high-frequency image component corresponding to each of the plurality of images acquired by the acquiring step;
A selection step of selecting any one of a plurality of parameter values detected by the detection step, and a parameter value selected by the selection step and one or more parameter values remaining after selection of the selection step An external control program for causing the processor to execute a normalizing step for normalizing a difference from each in cooperation with the internal control program. An image processing apparatus comprising: a capturing unit that captures an external control program; and a processor that executes processing according to the external control program captured by the capturing unit and an internal control program stored in a memory,
The external control program is
An acquisition step of repeatedly acquiring images capturing a common subject;
A detecting step for detecting a parameter value correlated with the amount of the high-frequency image component corresponding to each of the plurality of images acquired by the acquiring step;
A selection step of selecting any one of a plurality of parameter values detected by the detection step, and a parameter value selected by the selection step and one or more parameter values remaining after selection of the selection step An image processing apparatus corresponding to a program that executes a normalizing step for normalizing a difference from each of the two in cooperation with the internal control program.

Images