JP2010258885A - Image capturing apparatus and image processing method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image capturing apparatus capable of performing alignment processing by detecting shift between images when synthesizing a dynamic-range enlarged image from a plurality of images. <P>SOLUTION: An image capturing apparatus includes: a first image information capturing means 106 for capturing image information; an image specifying means 105 which calculates a feature value of image information from the image information captured by the first image information capturing means and specifies a partial region representing image features based on the image feature value; a second image information capturing means 101 for capturing image information; an image shift means 102 which inputs a plurality of image information items having different lengths of exposure time by means of the second image information capturing means and uses image partial region information specified by the image specifying means to detect a shift amount between images; a correction means 103 for performing inter-image shift correction processing according to the inter-image shift amount detected by the image shift means; and an image synthesis processing means 104 for synthesizing a wide dynamic-range image while using the images on which the shift correction processing has been performed by the correction means. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to an imaging apparatus and an image processing method.

  In general, in a photographing apparatus such as a digital camera, a solid-state image pickup device such as a CCD image pickup device is used. However, there is a problem that the dynamic range of the solid-state image pickup device is extremely narrow compared to a silver salt photographic film. Therefore, various techniques have been developed for expanding the dynamic range of the photographing apparatus.

  For example, as disclosed in Patent Documents 1 and 2, a plurality of shootings with different exposure amounts are performed on the same subject, that is, shooting with long exposure and short exposure is performed, and a normal processing is performed. There is a method of expanding the dynamic range as compared with an image.

  The above-described dynamic range expansion method by multiple shooting has an effect of expanding the dynamic range, but since multiple shooting is performed, care must be taken not to cause camera shake during shooting. When camera shake occurs, the synthesized image becomes a double image. If the camera is fixed using a tripod or the like, it is possible to prevent camera shake by preventing camera shake while expanding the dynamic range, but often it is not possible to use a tripod.

  Therefore, a method of measuring the amount of displacement between images and combining the images after alignment between images is also conceivable, but when measurement and displacement correction processing between images are performed inside the imaging device after being obtained, There exists a subject that processing time becomes long.

  It is an object of the present invention to provide an image processing method and an imaging apparatus that can detect a shift between images and perform alignment processing at a high speed when synthesizing a dynamic range expanded image of a plurality of images.

  In order to achieve the above object, in the photographing apparatus according to the present invention, a first image information acquisition unit for acquiring image information, and a feature amount of the image information from the image information acquired by the first image information acquisition unit, The length of the exposure time differs depending on the image specifying means for specifying the partial region that represents the image feature based on the image feature amount, the second image information acquiring means for acquiring the image information, and the second image information acquiring means. Image position shift means for detecting the shift amount between images using the image partial area information specified by the image specifying means by inputting multiple image information, and between the images based on the shift amount between images detected by the image position shift means A correction unit that performs a shift correction process and an image synthesis process unit that performs a synthesis process on a wide dynamic range image using the image that has been corrected by the correction unit.

  In the photographing apparatus according to the present invention, when the feature amount of the image information acquired by the first image information acquisition unit is the contrast of the image, the image specifying unit divides the image information into blocks, and the average value of the contrast for each block The image partial area for specifying the block position having a high contrast average value is specified, and the correction means detects and corrects the shift amount between the images using the partial area information specified by the image specifying means.

  In the photographing apparatus according to the present invention, when the feature amount of the image information acquired by the first image information acquisition unit is a face region of a person shown in the image, the image specifying unit displays the facial region as a partial region that represents the feature of the image. The correcting means detects and corrects the amount of deviation between images using the partial area information specified by the image specifying means.

  An image processing method according to the present invention includes: a first image information acquisition step for acquiring image information; a feature amount of the image information is obtained from the image information acquired in the first image information acquisition step; Input a plurality of pieces of image information having different exposure time lengths by an image specifying step for specifying a partial region that expresses the feature of the image, a second image information acquiring step for acquiring image information, and a second image information acquiring step, An image position shift step for detecting a shift amount between images using the image partial area information specified in the image specifying step, and a correction step for performing an image shift correction process based on the shift amount between images detected in the image position shift step. And an image synthesizing process step for synthesizing a wide dynamic range image using the image subjected to the shift correction process in the correcting step.

  In the image processing method according to the present invention, when the feature amount of the image information acquired in the first image information acquisition step is the contrast of the image, in the image specifying step, the image information is divided into blocks, and the average of the contrast for each block A value is obtained, and an image partial region that specifies a block position having a high contrast average value is specified. In the correction step, the amount of deviation between images is detected and corrected using the partial region information specified in the image specifying step.

  In the image processing method according to the present invention, when the feature amount of the image information acquired in the first image information acquisition step is a face region of a person shown in the image, in the image specifying step, the portion indicating the feature of the image in the face region In the correction step, the shift amount between images is detected and corrected using the partial region information specified in the image specifying step.

  An image forming program according to the present invention causes a computer to execute any one of the image processing methods described above, and a storage medium for recording the computer-readable image processing program according to the present invention, The image processing program is recorded.

  In the image processing method according to the present invention, an image for detecting the feature of the image is input by the first image information acquisition means, and the input image is reduced. Next, image features are extracted from the reduced image. The feature amount is an image contrast or a human face. For example, the image is divided into blocks, the average value of the contrast value is obtained in each block, and the average value of the block coordinates and the contrast of each block is output. As another example, the face area of a person is extracted from the image, and the coordinates of the face area are output.

  In the present invention, a plurality of images having different exposure times by the second image information acquisition means are input. Here, an image having a long exposure time and an image having a short exposure time are input. An inter-image shift amount is measured using the image feature amount obtained by extracting the image feature from the reduced image. Here, the average contrast values for each block of the divided image are arranged in the descending order, and a small number of blocks with high contrast are selected. Detection of the shift of the image input by the second image information acquisition means in the vicinity of the coordinates of the selected block is performed. The selected block occupies only a part of the full screen of the image. Since this part is a block with high contrast, effective displacement measurement can be performed by using these blocks. In addition, since it is not necessary to detect deviation for all images, the processing time can be shortened.

  In a shooting device with a face detection mode, that is, a shooting device that detects the face area and can adjust the focus, brightness, and color according to the face area, instead of detecting a block with high contrast, the human face area is detected from the image. Detect and measure the amount of deviation of the image input by the second image information acquisition means in the vicinity of the face area. Since the face area is only a part of the entire area of the image, the processing time for measuring the amount of deviation near the face area can be significantly reduced compared to the detection of deviation on the entire screen. Since the face area is the attention area, if the misalignment correction is performed in accordance with the face area, the misalignment correction can be performed accurately and accurately on the face area even if there is a misalignment correction error of the entire screen. In addition, by using the detected shift amount between images, shift correction between images is performed, and a wide dynamic range image is synthesized.

  According to the present invention, an image for detecting the feature of the image is input by the first image information acquisition means for acquiring the image information, and the feature amount is detected from the image by the image specifying means. Since a characteristic area is extracted, the characteristic area occupies only a part of the entire image, and displacement detection in this area can be effectively performed. For this reason, when synthesizing a dynamic range expanded image of a plurality of images, alignment processing between images can be performed at high speed, and processing time can be shortened.

  According to the present invention, the second image information acquisition unit that acquires image information inputs a plurality of pieces of image information of an image having a long exposure time and an image having a short exposure time, and is extracted by the first image information acquisition unit. In the vicinity of the feature region, a shift amount between images is obtained to perform shift correction. For this reason, by correcting the shift between images, a double image can be eliminated when a dynamic range enlarged image is synthesized. Also, when setting the face detection, if the face area detected from the image is the feature area, the face area is only a part of the whole screen. When synthesizing a range-enlarged image, alignment processing between images can be performed at high speed, and processing time can be shortened.

  For this reason, effective and high-speed image misalignment detection and misalignment correction can be performed, so when synthesizing a wide dynamic range expanded image, the double image of the synthesized image can be eliminated and a dynamic range expanded image composed of a plurality of images can be synthesized. In this case, it is possible to provide an image processing method and an imaging apparatus that can detect a shift between images and perform alignment processing at high speed.

It is a block diagram which shows the structure of the image process part contained in the imaging device concerning 1st embodiment of this invention. It is a block diagram which shows the example of 1 structure of an image feature detection part. It is a block diagram for demonstrating the state which divided | segmented the image at the time of extracting the contrast information of an image as a feature-value in an image feature detection part, and the processing content. It is a figure for demonstrating the example of the edge extraction by the edge extraction filter in an image feature detection part. It is a figure which shows the edge image after an edge filter process. (A) is a diagram showing a selected high-contrast block position in one image input by the second image input unit, and (b) is a high contrast in another image input by the second image input unit. It is the figure which showed the block position. It is a figure which shows the weight function used for an image composition in an image composition process part. It is a block diagram which shows the structure of the imaging device to which 1st embodiment was applied. It is a flowchart which shows the processing content regarding the image shift detection, shift correction, and image composition which are performed inside the imaging device shown in FIG. It is a block diagram which shows the structure of the image process part contained in the imaging device concerning 2nd embodiment of this invention. (A) is the figure which showed the position of the selected face area in one image input with the 2nd image input part, (b) is the position of the face area in another image input with the 2nd image input part. FIG. It is a block diagram which shows the structure of the imaging device to which 2nd embodiment was applied. FIG. 13 is a flowchart showing processing contents related to face area recognition, image shift correction, and image composition executed in the imaging apparatus shown in FIG. 12. FIG.

  DESCRIPTION OF EMBODIMENTS Hereinafter, embodiments for carrying out an imaging apparatus, an image processing method, and an image processing program according to the present invention will be described in detail with reference to the drawings. In the drawings, the same or corresponding parts are denoted by the same reference numerals, and redundant description thereof is simplified or omitted.

In addition, an image processing program according to the present invention is installed and stored in advance in each imaging apparatus described below. The image processing program is readable by a computer, stored in a well-known optical storage medium such as a CD or a DVD (not shown), and activated by being installed in a storage unit of each imaging apparatus. The image processing method according to the present invention is configured to be executed.
(First embodiment)
A configuration example of the image processing unit 100 provided in the digital camera 200 serving as a photographing apparatus to which the present invention is applied will be described. FIG. 1 is a block diagram showing a configuration of an image processing unit 100 according to the present embodiment.

  In FIG. 1, an image processing unit 100 includes a second image input unit 101 serving as a second image information acquisition unit, an image shift detection unit 102 serving as an image position shift unit, an image shift correction unit 103 serving as a correction unit, and an image. An image composition processing unit 104 serving as a composition processing unit, an image feature detection unit 105 serving as an image specifying unit, a first image input unit 106 serving as a first image information acquisition unit, and an image output unit 107 are provided.

  The second image input unit 101 inputs an image from the digital camera 200, and inputs a plurality of pieces of image information having different exposure times for synthesizing a wide dynamic range image. In this case, information on each of an image with a long exposure time and a short image is input.

  The first image input unit 106 is for shooting the same shooting target as the second image input unit 101 and inputting the image. Details of the configuration and processing of the image feature detection unit 105 will be described with reference to FIG. The image feature detection unit 105 reduces the image input from the first image input unit 106 by the reduced image generation unit 201, and an image feature information extraction unit 202 uses the feature amount of the reduced image as feature information. Extract using. In this embodiment, the image feature information extraction unit 202 extracts image contrast information, that is, edge information as a feature amount. As an extraction method, edge extraction filter processing using an edge extraction filter is performed on the reduced image. In this embodiment, as an example of edge extraction, a well-known Laplacian filter serving as the edge extraction filter of Table 1 is used.

  In this embodiment, the edge information (image information) extracted by the edge extraction filter used in the edge extraction filter processing is divided into blocks as shown in FIG. 3, and the edge information after the edge extraction filter processing is obtained in each block. Performs averaging processing within the block. In this embodiment, an example of a Laplacian filter has been described as an example of an edge extraction filter, but other edge extraction filters may be used.

  Next, an example of edge extraction by the edge extraction filter will be described.

  FIG. 4 is a diagram in which the horizontal axis represents the x coordinate and the vertical axis represents the luminance value. When an edge is extracted from an image having characteristics as shown in FIG. 4, an edge as shown in FIG. 5 is obtained. That is, in FIG. 4, a portion where the luminance value is bent and changes is regarded as an edge. The average value of this edge is shown in FIG. In FIG. 3, dark blocks indicate high contrast blocks.

  The image shift detection unit 102 illustrated in FIG. 1 detects the shift amount between images with respect to a plurality of images input by the second image input unit 101. Here, first, the image feature detection result shown in FIG. 3 input from the image feature detection unit 105 is input. Based on the result, a block having a high contrast average value (a dark block) is selected. The selected high-contrast block position is indicated by black blocks A, B, C, and D in the image 1 shown in FIG. 6A input by the second image input unit 101. The position of the corresponding block on the screen of another image 2 input by the second image input unit 101 is changed to the black blocks A ′, B ′, C ′, D ′ in the image 2 of FIG. It shows with.

  Next, the amount of shift between images is detected near a black block with high contrast. If the black blocks A, B, C, and D with high contrast in the image 1 in FIG. 6A are set as the start positions, the blocks A ′, B ′, and C in the image 2 in FIG. The amount of deviation of “, D” is detected.

  This deviation amount detection method is shown in Equation 1. The black block size with high contrast of the image 1 in FIG. Using the pixel value f (i, j) of the image 1 in the region and the pixel value g (i, j) of the image 2, the value of the following equation 1 is calculated. Here, (i, j) is the coordinate value of the pixel on the image 1. g (i, j) is a coordinate value of a pixel on the image 2. The position (u, v) when E (u, v) is minimum is obtained. This value is the shift amount of the corresponding block between image 1 and image 2.

  The total shift amount between the image 2 and the image 1 is calculated from the calculated shift amount of each black block. The average shift and rotation amount of the image is determined from the shift amount of each block, and the shift amount of the entire image is determined.

  The image shift correction unit 103 corrects a shift between images 1 and 2. Here, the amount of deviation between images detected by the image deviation detection unit 102 is used, and the image 2 is entirely corrected by the amount of deviation. In other words, the shift between the image 1 and the image 2 is corrected using the parallel movement and the rotation amount of the image detected by the image shift detection unit 102. As a correction method, the shift of the image 2 is corrected using the translation amount and the rotation amount of the image 2 measured.

  The image composition processing unit 104 performs dynamic range expansion image composition processing after correcting the image shift. Here, a composite example of an image shot with a long exposure time and an image shot with a short exposure time will be described. FIG. 7 shows an image composition weighting function, the horizontal axis represents the luminance value of the image, and the vertical axis represents the weighting coefficient for image composition.

When synthesizing images having different exposure times, first, normalization of images having a short exposure time is performed by the following formula 2. Here, I ′ _S is the normalized pixel value of the short exposure image, and I _S is the pixel value before normalization. K is a normalization factor, which is the ratio of long exposure time to short exposure time.

Next, the normalized image is synthesized using a weight function shown in FIG. As shown in FIG. 7, when the luminance value of the pixel is larger than M2, the pixel value of an image having a short exposure time is selected. If it is smaller than the brightness knowledge M1, a pixel value of an image having a long exposure time is selected. If the luminance value is between M1 and M2, the pixel value is obtained by a weighted average of two images. The weighted average weighting factor uses the weighting factor in the hatched portion between M1 and M2 shown in FIG. Luminance knowledge M1 and M2 are determined in advance through experiments. In this manner, the image composition processing unit 104 synthesizes a plurality of images for dynamic range expansion.
An image output unit 107 shown in FIG. 1 outputs a processed image and other images.

Next, a hardware configuration of the digital camera 200 that performs the above-described image processing will be described.
FIG. 8 is a block diagram showing a hardware configuration of the digital camera 200 according to the present embodiment. As shown in FIG. 8, the digital camera 200 includes a photographing optical system 1, a mechanical shutter 2, a CCD (Charge Coupled Device) 3, a CDS (Correlated Double Sampling) circuit 4, an A / D converter 5, and a drive. Motor 6 as a source, timing signal generator 7 for generating a timing signal, signal processing circuit 8, CPU (Central Processing Unit) 9, RAM (Random Access Memory) 10, ROM (Read Only Memory) 11, SDRAM (Synchronous DRAM) 12, an image compression / decompression circuit 13, a memory card 14, an operation unit 15, and a liquid crystal display 16 (hereinafter referred to as “LCD 16”).

  The subject light L is incident on the CCD 3 through the photographing optical system 1. A mechanical shutter 2 is disposed between the photographing optical system 1 and the CCD 3, and incident light to the CCD 3 can be blocked by the mechanical shutter 2. The photographing optical system 1 and the mechanical shutter 2 are each driven by a motor 6.

  The CCD 3 converts an optical image formed on the imaging surface into an electrical signal and outputs it as analog image data. The image information output from the CCD 3 is subjected to noise component removal by the CDS circuit 4, converted into a digital value by the A / D converter 5, and then output to the image processing circuit 8.

  The signal processing circuit 8 has a function of performing various image processing such as YCrCb conversion processing, white balance control processing, contrast correction processing, edge enhancement processing, and color conversion processing using the SDRAM 12 that temporarily stores image data. The white balance process is an image process that adjusts the color density of image information, and the contrast correction process is an image process that adjusts the contrast of image information. The edge enhancement process adjusts the sharpness of image information, and the color conversion process is a well-known image process that adjusts the hue of image information. The image processing circuit 8 displays image information subjected to signal processing and image processing on a liquid crystal display 16 (hereinafter abbreviated as “LCD 16”). The image information subjected to signal processing and image processing is recorded in the memory card 14 via the image compression / decompression circuit 13. The image compression / decompression circuit 13 compresses the image information output from the image processing circuit 8 according to the instruction acquired from the operation unit 15 and outputs the compressed image information to the memory card 14, and decompresses the image information read from the memory card 14. It is a circuit that outputs to the signal processing circuit 8.

  The timing of the CCD 3, the CDS circuit 4, and the A / D converter 5 is controlled by the CPU 9 via the timing signal generator 7. Further, the image processing circuit 8, the image compression / decompression circuit 13, and the memory card 14 are also controlled by the CPU 9.

  In the digital camera 200, the CPU 9 performs various arithmetic processes according to the program, stores the various programs and the like in the ROM 11 that is a read-only memory serving as a storage unit, and uses the various programs in various processing steps that the RAM 10 that is a read / write memory has. Various data is developed in the work area and various data storage areas. The image processing circuit 8 to the operation unit 15 are connected to be communicable with each other via a bus line. In this embodiment, the image processing program is stored in the ROM 11 before product shipment. Of course, when upgrading the program, the latest image processing program is installed in the ROM 11 after the product is shipped.

  Processing steps by the digital camera 200 having such a configuration will be described with reference to the flowchart shown in FIG.

  First, in the digital camera 200, image information for acquiring image feature information is input by the first image input unit 101, and the image information reduced through the image processing circuit 8 is temporarily stored in the memory area of the SDRAM 12. The reduced image data in the memory area is obtained again by calculating the average contrast value for each block of the image by the image processing circuit 8, and the calculated average contrast value for each block is stored in the image memory area of the SDRAM 12. Thereby, the block number and the contrast average value for each block are stored.

  Next, actual shooting is performed. Here, two images having different image exposure amounts are photographed, and the photographed image input from the second image input unit 101 is temporarily stored in the SDRAM 12.

  In the digital camera 200, an image shift detection program, which is one of image processing programs stored in advance in the ROM 11, is activated to perform shift detection processing. The image shift detection program selects a plurality of blocks having the highest contrast value from the block number stored in the SDRAM 12 and the contrast average value for each block. At the position of the selected block, the shift amount is detected from the image information of a plurality of main shooting images stored in the SDRAM 12. That is, the image shift amount is obtained, and the average movement coefficient and rotation coefficient of the image are obtained from the block shift amount.

Next, the translation coefficient and rotation coefficient between images obtained from the obtained deviation amount and the image synthesis weight coefficient parameter (see FIG. 7) stored in the ROM 11 are called, and images with different exposure times stored in the SDRAM 12 are retrieved. The information is transferred to the image processing circuit 8 together with the information, and the image processing circuit 8 corrects the shift between the images by parallel movement and rotation processing of the image, and performs wide dynamic range image synthesis processing by the synthesis weight coefficient. An image that has been subjected to various types of processing is displayed on the LCD 16 and is output as an image.
(Second Embodiment)
When the digital camera 201 serving as the photographing apparatus according to the present embodiment detects a human face and inputs an image to synthesize a dynamic range expanded image, the digital camera 201 detects a positional shift between the images at a high speed and detects the image shift. It is to correct.

  A configuration example of the image processing unit 100A included in the digital camera 201 according to the second embodiment will be described with reference to FIG. FIG. 10 is a block diagram showing the configuration of the image processing unit 100A according to this embodiment. The image processing unit 100A, together with the second image input unit 101, the image shift detection unit 102, the image shift correction unit 103, the image composition processing unit 104, the first image input unit 106, and the image output unit 107, here, A face area detection unit 105A is provided as a feature detection unit.

  The second image input unit 101 inputs an image from the digital camera 201, and inputs a plurality of pieces of image information having different exposure times for synthesizing a wide dynamic range image. Here, information on each of an image having a long exposure time and an image having a short exposure time is input.

  The first image input unit 106 shoots the same shooting target as the second image input unit 101 and inputs an image. The image face area detection unit 105A has a function of reducing an input image, detecting a human face area with respect to the reduced image, and outputting coordinates of the detected face area. In addition, since the structure which detects a human face is already well-known, the description is abbreviate | omitted here.

  The image shift detection unit 102 detects a shift amount between images for a plurality of pieces of image information input by the second image input unit 101. Here, first, the face area detection result input from the face area detection unit 105A is input. Using the result, the position of the face area shown in image 1 of FIG. 11A and the corresponding face area on the screen of another image 2 input by the second image input unit 101 is shown in FIG. The image 2 shown in FIG.

  Next, image misalignment detection is performed in the vicinity of the detected face area of image 1. With the face area of image 1 in FIG. 11A as the start position, the amount of deviation in image 2 in FIG. 11B is detected by the following deviation detection method.

  The above equation 1 is used as the deviation amount detection method. Here, the face area size of the image 1 is M × N. The value of Equation 1 is calculated using the pixel value f (i, j) of the image 1 in the region and the pixel value g (i, j) of the image 2 to be shifted. Here, (i, j) is the coordinate value of the pixel on the image 1. The position (u, v) when E (u, v) is minimum is obtained. This value is the amount of deviation between image 1 and image 2.

  From the obtained shift amount of each face area, the total shift amount between the image 2 and the image 1 is calculated. In the case where there are a plurality of images, the shift between the images can be obtained by averaging the shift amounts of the face areas.

  After obtaining the image shift amount, the image shift correction and image composition processing are the same as those in the first embodiment.

  Next, a hardware configuration of the digital camera 201 that performs the above-described image processing will be described. FIG. 12 is a block diagram showing a hardware configuration of the digital camera 201 according to the present embodiment. As shown in FIG. 12, the digital camera 201 includes a photographing optical system 1, a mechanical shutter 2, a CCD 3, a CDS circuit 4, an A / D converter 5, a motor 6 as a driving source, and a timing signal generator 7 that generates a timing signal. , A signal processing circuit 8, a CPU 9, a RAM 10, a ROM 11, an SDRAM 12, an image compression / decompression circuit 13, a memory card 14, an operation unit 15, and an LCD 16.

  Then, the subject light L1 first enters the CCD 3 through the photographing optical system 1. A mechanical shutter 2 is disposed between the photographing optical system 1 and the CCD 3, and incident light on the CCD 3 can be blocked by the mechanical shutter 2. The photographing optical system 1 and the mechanical shutter 2 are driven by a motor 6.

  The CCD 3 outputs an optical image formed on the imaging surface as analog image data. The image information output from the CCD 3 is subjected to noise component removal by the CDS circuit 4, converted into a digital value by the A / D converter 5, and then output to the image processing circuit 8.

The image processing circuit 8 performs various image processing such as YCrCb conversion processing, white balance control processing, contrast correction processing, edge enhancement processing, and color conversion processing using the SDRAM 12 that temporarily stores image information. The white balance process is an image process that adjusts the color density of image information, and the contrast correction process is an image process that adjusts the contrast of image information. The edge enhancement process adjusts the sharpness of image information, and the color conversion process is an image process that adjusts the hue of image information. The image processing circuit 8 displays on the LCD 16 image information that has undergone signal processing and image processing. The image information subjected to signal processing and image processing is recorded in the memory card 14 via the image compression / decompression circuit 13. The image compression / decompression circuit 13 compresses the image information output from the image processing circuit 8 according to the instruction acquired from the operation unit 15 and outputs the compressed image information to the memory card 14, and decompresses the image information read from the memory card 14. It is a circuit that outputs to the signal processing circuit 8.
The timing of the CCD 3, the CDS circuit 4, and the A / D converter 5 is controlled by the CPU 9 via the timing signal generator 7. Further, the image processing circuit 8, the image compression / decompression circuit 13, and the memory card 14 are also controlled by the CPU 9.

  In the digital camera 201, the CPU 9 performs various arithmetic processes according to the program, stores the various programs in the ROM 11 that is a read-only memory serving as a storage unit, and uses the various programs in various processing processes of the RAM 10 that is a readable / writable memory. Various data is developed in the work area and various data storage areas. The image processing circuit 8 to the operation unit 15 are connected to be communicable with each other via a bus line. In this embodiment, the image processing program is stored in the ROM 11 before product shipment. Of course, when upgrading the program, the latest image processing program is installed in the ROM 11 after the product is shipped.

  Processing steps by the digital camera 201 having such a configuration will be described with reference to a flowchart shown in FIG.

  In the digital camera 201, when image data for detecting a face area is input by the first image input unit 101, the reduced image data is temporarily stored in the memory area SDRAM 12 through the image processing input circuit 8. Next, the face detection program 11 constituting the image processing program stored in advance in the ROM 11 is activated to perform face detection processing. The face detection result, that is, the coordinate values A, B, C, and D shown in FIG. 11A indicating the detected face area are stored in the SDRAM 12.

  Next, the second image input unit 101 performs main shooting, two images having different exposure amounts are shot, and the shot image data is temporarily stored in the SDRAM 12. Next, an image shift detection program constituting an image processing program stored in advance in the ROM 11 is activated to perform shift detection processing. In the shift detection program, the coordinate values A, B, C, D of the face area stored in the SDRAM 12 are input. Then, using the face area position data, the amount of deviation is detected from the image stored in the SDRAM 12.

  Next, an image composition weighting coefficient parameter (see FIG. 7) stored in the ROM 11 is called up and the image processing circuit 8 together with images of different exposure times in the SDRAM 12, the obtained image displacement amount and weighting coefficient are called. And image misalignment correction and image composition processing are performed inside the image processing circuit 8. An image that has been subjected to various types of processing is displayed on the LCD 16 and is output as an image.

  As described above, in each of the above embodiments, an image for detecting image features is input by the first image input unit 106 that acquires image information, and the feature amount is detected from the image by the image feature detection unit 105. Since the most characteristic area is extracted on the screen, the characteristic area occupies only a part of the entire image, and displacement detection in this area can be effectively performed. For this reason, when synthesizing a dynamic range expanded image of a plurality of images, alignment processing between images can be performed at high speed, and processing time can be shortened.

  The second image input unit 101 that acquires image information inputs a plurality of pieces of image information of an image having a long exposure time and an image having a short exposure time, and the feature region extracted by the first image input unit 106 is input. In the vicinity, the image shift detection unit 102 and the image shift correction unit 103 obtain a shift amount between images and perform shift correction. For this reason, by correcting the shift between the images, the double image can be eliminated when the image synthesis processing unit 104 synthesizes the dynamic range expanded image.

  For this reason, effective and high-speed image misalignment detection and misalignment correction can be performed, so when synthesizing a wide dynamic range expanded image, the double image of the synthesized image can be eliminated and a dynamic range expanded image composed of a plurality of images can be synthesized. In this case, it is possible to provide an image processing method and an imaging apparatus that can detect a shift between images and perform alignment processing at high speed.

  Further, as in the second embodiment, when setting face detection, if the face area detected from the image is a feature area, the face area is only part of the entire screen, and thus detection of deviation near the face area is performed. If the processing is performed, when a dynamic range expanded image composed of a plurality of images is synthesized, alignment processing between the images can be performed at high speed, and the processing time can be shortened.

DESCRIPTION OF SYMBOLS 101 Second image information acquisition means 102 Image position shift means 103 Correction means 104 Image composition processing means 105 Image identification means 106 First image information acquisition means 200, 201 Image capturing apparatus 300 Storage medium

JP 2000-92378 A Japanese Patent No. 3974799

Claims (8)

First image information acquisition means for acquiring image information;
An image specifying means for determining a feature amount of the image information from the image information acquired by the first image information acquiring means, and specifying a partial region representing the feature of the image by the feature amount of the image;
Second image information acquisition means for acquiring image information;
Image position shift in which a plurality of pieces of image information having different exposure times are input by the second image information acquisition unit, and the amount of shift between images is detected using the image partial area information specified by the image specifying unit Means,
Correction means for performing an inter-image shift correction process based on a shift amount between images detected by the image position shift means;
An image capturing apparatus comprising: an image composition processing unit configured to perform composition processing of a wide dynamic range image using an image subjected to shift correction processing by the correction unit. The feature amount of the image information acquired by the first image information acquisition means is the contrast of the image,
The image specifying unit divides the image information into blocks, obtains an average value of contrast for each block, specifies an image partial region that specifies a block position having a high contrast average value,
The imaging apparatus according to claim 1, wherein the correction unit detects and corrects a shift amount between images using the partial region information specified by the image specifying unit. The feature amount of the image information acquired by the first image information acquisition means is a face area of a person shown in the image,
The image specifying means sets the face area as a partial area representing the characteristics of an image,
The imaging apparatus according to claim 1, wherein the correction unit detects and corrects a shift amount between images using the partial region information specified by the image specifying unit. A first image information acquisition step of acquiring image information;
An image specifying step for obtaining a feature amount of the image information from the image information acquired in the first image information acquisition step, and specifying a partial region representing the image feature based on the image feature amount;
A second image information acquisition step of acquiring image information;
Image position shift step of inputting a plurality of pieces of image information having different exposure times in the second image information acquisition step and detecting a shift amount between the images using the image partial area information specified in the image specifying step When,
A correction step of performing an inter-image shift correction process based on the shift amount between the images detected in the image position shift step;
An image processing method comprising: an image synthesis processing step of synthesizing a wide dynamic range image using the image subjected to the shift correction processing in the correction step. The feature amount of the image information acquired in the first image information acquisition step is an image contrast,
The image specifying step divides the image information into blocks, obtains an average value of contrast for each block, specifies an image partial region that specifies a block position having a high contrast average value,
The image processing method according to claim 4, wherein the correcting step detects and corrects a shift amount between images using the partial region information specified in the image specifying step. The feature amount of the image information acquired in the first image information acquisition step is a face area of a person shown in the image,
In the image specifying step, the face area is set as a partial area representing the characteristics of the image,
The image processing method according to claim 4, wherein the correcting step detects and corrects a shift amount between images using the partial region information specified in the image specifying step.   An image forming program for causing a computer to execute the image processing method according to claim 4. A storage medium for recording a computer-readable image processing program,
8. A storage medium in which the image processing program according to claim 7 is recorded as the image processing program.

Images