JP2009200923A - Image processing apparatus and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image processing apparatus and a program for relatively reducing a processing load in extracting a region of a corrected portion from a first image scanned from a document and from a second image scanned from a corrected document of the document. <P>SOLUTION: The image processing apparatus includes: a reduction unit 62 for reducing, in a predetermined reduction scale, a first image scanned from a document and a second image scanned from a corrected document obtained by correcting the document; a differential image generation unit 64 for generating a differential image between the reduced first image and the reduced second image; a merged region determination unit 66 for determining a merged region, that is a region in the differential image, meeting predetermined conditions for significant pixel distribution; and a corrected portion extraction unit 68 for extracting a region of a corrected portion based on comparison of an image in a region corresponding to the merged region in the first image and an image in a region corresponding to the merged region in the second image. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to an image processing apparatus and a program.

There is a technique for extracting a region of a corrected portion based on a document image scanned from a document and a modified document image scanned from the modified document of the document. For example, in Patent Document 1 below, a difference between a document image and a corrected document image is generated, and an area in which the document image and the corrected document image are compared in detail is specified based on the difference. Then, the area of the corrected portion is extracted by comparing the original image in the specified area with the corrected original image in the area in detail.
JP 2007-241356 A

  However, in the above-described conventional technique, depending on the sizes of the original image and the corrected original image, the processing load for generating the difference between the original image and the corrected original increases, and as a result, it takes time to extract the area of the corrected portion. There was such a case.

  The present invention has been made in view of the above problems, and an object of the present invention is to extract a region of a corrected portion from a first image scanned from a document and a second image scanned from the modified document of the document. It is an object to provide an image processing apparatus and a program that can reduce the processing load at the time.

  According to a first aspect of the present invention, there is provided a reduction means for reducing the first image scanned from the original and the second image scanned from the corrected original corrected on the original at a predetermined reduction ratio, and the reduced first image. A difference image generation means for generating a difference image between the first image and the reduced second image; an area determination means for determining an area in the difference image that satisfies a predetermined condition regarding a significant pixel distribution; A corrected portion that compares the image of the corresponding region of the determined region in one image with the image of the corresponding region of the determined region in the second image, and extracts the region of the corrected portion based on the comparison result And an extracting unit.

  According to a second aspect of the present invention, in the first aspect of the invention, the reduced first image, feature extraction means for extracting feature points of the reduced second image based on significant pixels, and the extracted feature points And calculating means for calculating a correction parameter for correcting a positional deviation between the reduced first image and the reduced second image, and the difference image generating means is configured to calculate the reduced first image according to the correction parameter. The difference image is generated after correcting a positional deviation between the image and the reduced second image.

  According to a third aspect of the present invention, in the second aspect of the invention, the region image of the inspection region that occupies a part or all of the first image or the second image is reduced at the predetermined reduction rate while changing a reduction condition. A temporary reduction means, a comparison means for comparing the area image and the reduced area image based on significant pixels, and a selection means for selecting one reduction condition based on a result of comparison by the comparison means; , And the reduction means reduces the first image and the second image under the selected reduction condition.

  According to a fourth aspect of the present invention, in the third aspect of the invention, the comparison unit calculates a numerical value corresponding to a ratio between the number of significant pixels of the region image and the number of significant pixels of the reduced region image. The selection means selects a reduction condition when the numerical value becomes a value closest to a target value.

  According to a fifth aspect of the present invention, in the third aspect of the invention, the comparison means extracts a feature point of the region image and the reduced region image based on significant pixels, and a feature point of the region image. And a means for counting the number of feature points also present in the reduced area image, wherein the selection means selects a reduction condition when the number is the largest.

  According to a sixth aspect of the present invention, in the second aspect of the present invention, the region image of the inspection region occupying part or all of the first image or the second image is reduced at the predetermined reduction rate while changing a reduction condition. Temporary reduction means, enlargement means for enlarging the reduced area image to a size before reduction, comparison means for comparing the enlarged reduced area image with the area image based on significant pixels, Selecting means for selecting one reduction condition based on the result of the comparison by the comparison means, wherein the reduction means includes the first image and the second image under the selected reduction condition. It is characterized by reducing.

  According to a seventh aspect of the invention, in the sixth aspect of the invention, the comparing means performs a logical operation between the pixels corresponding to the positions, whereby the degree of coincidence between the enlarged reduced area image and the area image. The selection means calculates a reduction condition when the numerical value becomes a peak value.

  According to an eighth aspect of the present invention, in the invention according to any one of the second to seventh aspects, the correction portion extraction unit is configured to determine the image of the corresponding region in the first image and the second image based on the correction parameter. After the alignment with the image of the corresponding area, the image of the corresponding area in the first image and the image of the corresponding area in the second image are compared.

  According to a ninth aspect of the present invention, in any one of the third to eighth aspects, the reduction condition is an image reduction method.

  According to a tenth aspect of the present invention, in the invention according to any one of the third to eighth aspects, the reduction means performs OR reduction of the first image and the second image, and the temporary reduction means includes the region image. , And the reduction condition is a threshold pixel number for OR reduction.

  According to an eleventh aspect of the present invention, there is provided a reduction means for reducing a first image scanned from a document and a second image scanned from a modified document in which the document has been corrected, a reduced first image, and a reduced first image. A difference image generating means for generating a difference image with the reduced second image, an area determination means for determining an area in the difference image that satisfies a predetermined condition regarding a significant pixel distribution, As a corrected portion extracting means for comparing the image of the corresponding region of the determined region with the image of the corresponding region of the determined region in the second image, and extracting the region of the corrected portion based on the comparison result A program that causes a computer to function.

  In the first and eleventh aspects of the invention, an area (corresponding area) for comparing the first image and the second image is specified based on the difference image between the reduced first image and the reduced second image. Therefore, for example, the processing load for specifying the corresponding region is relatively less than when the corresponding region is specified based on the difference image between the first image and the second image. As a result, according to the inventions of claims 1 and 11, it is possible to relatively reduce the processing load when extracting the corrected portion.

  In the invention of claim 2, for example, the corresponding region is specified after correcting the positional deviation between the reduced first image and the reduced second image. As a result, the accuracy of specifying the corresponding area is improved, so that the extraction accuracy of the corrected portion can be improved.

  In the invention of claim 3, the area image of the inspection area occupying all or part of the first image or the second image is reduced while changing the reduction condition, and the images before and after the reduction are compared based on significant pixels. . Thereby, in the invention of claim 3, for example, it is determined whether or not the feature necessary for calculating the correction parameter is maintained after the reduction. In the third aspect of the invention, one reduction condition is selected based on the comparison result, and the first image and the second image are reduced under the reduction condition. Thus, in the invention of claim 3, for example, the first image and the second image are easily reduced under a reduction condition in which the features necessary for calculating the correction parameter are maintained as much as possible after the reduction. Therefore, according to the third aspect of the invention, since the accuracy of correcting the positional deviation is improved, the extraction accuracy of the corresponding region is also improved. As a result, there is a high possibility that the correction accuracy of the corrected portion is improved.

  In the invention of claim 4, for example, the first image and the first image are reduced under the reduction condition when the ratio of the number of significant pixels in the region image and the number of significant pixels in the reduced region image is closest to the target value. Reduce two images. Therefore, according to the fourth aspect of the present invention, the first image and the second image can be reduced under a reduction condition in which the feature necessary for calculating the correction parameter is most maintained after reduction. Therefore, since the accuracy of correction of the positional deviation is improved, the extraction accuracy of the corresponding region is also improved, and as a result, the possibility of improving the extraction accuracy of the corrected portion is increased.

  According to the invention of claim 5, it is possible to reduce the first image and the second image under a reduction condition in which the features necessary for calculating the correction parameter are most maintained after the reduction. Therefore, since the accuracy of correction of the positional deviation is improved, the extraction accuracy of the corresponding region is also improved, and as a result, the possibility of improving the extraction accuracy of the corrected portion is increased.

  According to the sixth aspect of the present invention, the first image and the second image can be easily reduced under the reduction condition that maintains the characteristics necessary for calculating the correction parameter as much as possible even after the reduction, as compared with the case without this configuration. . For this reason, the accuracy of correcting the positional deviation is improved, so that the extraction accuracy of the corresponding region is also improved. As a result, there is a high possibility that the extraction accuracy of the corrected portion is improved.

  According to the seventh aspect of the present invention, for example, the first image and the second image can be reduced under a reduction condition in which the feature necessary for calculating the correction parameter is most maintained after reduction. For this reason, the accuracy of correcting the positional deviation is improved, so that the extraction accuracy of the corresponding region is also improved. As a result, there is a high possibility that the extraction accuracy of the corrected portion is improved.

  According to the eighth aspect of the present invention, it is possible to improve the extraction accuracy of the corrected portion compared to the case where the present configuration is not provided.

  According to the ninth aspect of the present invention, it is possible to determine a reduction method that can extract the region of the corrected portion with high accuracy.

  According to the tenth aspect of the present invention, in the case of extracting the region of the corrected portion by OR reduction of the first image and the second image, the threshold pixel of the OR reduction that can extract the region of the corrected portion with high accuracy You can figure out the number.

  Embodiments of the present invention will be described with reference to the drawings. FIG. 1 is an example of a hardware configuration diagram of an image processing apparatus 2 according to the present invention. As shown in the figure, the image processing apparatus 2 includes a central processing unit 40, a main storage unit 42, an auxiliary storage unit 44, an operation unit 46, and a display 48. Each unit can communicate with each other via the data bus 50. The image processing apparatus 2 includes a network interface, a printer, a scanner, and the like (not shown).

  The central processing unit 40 is a CPU, MPU, or the like. The central processing unit 40 operates according to a program stored in advance in the main storage unit 42 and controls each unit. In addition, the central processing unit 40 calculates using information stored in the main storage unit 42 and outputs it to the main storage unit 42. The program is not limited to the one stored in the main storage unit 42 but may be stored in an information storage medium such as a CD-ROM or DVD-ROM, or supplied from a network. There may be.

  The main storage unit 42 is a memory element such as a RAM or a ROM. The main storage unit 42 stores the program. Further, the main storage unit 42 stores information input from each unit.

  The auxiliary storage unit 44 is a storage medium such as a hard disk. The auxiliary storage unit 44 stores information stored in the main storage unit 42 in accordance with a control signal from the central processing unit 40. In the present embodiment, the auxiliary storage unit 44 stores a first image scanned from a paper document and a second image scanned from a corrected document in which correction such as additional writing or whiteout is made on the paper document. is doing. The first image and the second image are images scanned by a scanner (not shown), for example.

  The operation unit 46 is a keyboard, a mouse, or the like. The operation unit 46 outputs user operation information to the main storage unit 42. The display 48 displays information stored in the main storage unit 42 in accordance with a control signal from the central processing unit 40.

  In the present embodiment, the image processing device 2 extracts and outputs the region of the correction portion based on the first image and the second image. FIG. 2 is a diagram illustrating a state in which the region of the correction portion is extracted based on the first image and the second image. FIG. 2A is an example of the first image. FIG. 2B is an example of the second image. A hatched portion indicates a corrected portion. FIG. 2C is a diagram illustrating a state where the region of the correction portion is extracted.

  In the image processing apparatus 2, the processing load when extracting the region of the corrected portion is intended to be relatively reduced. Hereinafter, a technique for extracting the region of the corrected portion with a relatively small processing load will be described.

  FIG. 3 is a block diagram of functions realized in the image processing apparatus 2 when the central processing unit 40 operates according to the program. Here, the functions related to the present invention are mainly shown. As shown in the figure, the image processing apparatus 2 includes a reduction condition determination unit 60, a reduction unit 62, a difference image generation unit 64, an integrated region determination unit 66, and a corrected portion extraction unit 68. The reduction condition determination unit 60 includes a temporary reduction unit 60a, a temporary comparison unit 60b, and a selection unit 60c. The corrected portion extraction unit 68 includes a position deviation correction unit 68a, an image division unit 68b, and a difference extraction unit 68c.

  The reduction condition determination unit 60 is realized with the central processing unit 40 as the center. The reduction condition determination unit 60 determines a reduction condition when the reduction unit 62 reduces the first image and the second image at a predetermined reduction rate. Here, the reduction condition is a known image reduction method such as a thinning method, a nearest neighbor method, or the like.

  The temporary reduction unit 60a reduces an area image of a predetermined inspection area that occupies a part or all of the first image at a predetermined reduction ratio while changing a reduction condition. In this way, the temporary reduction unit 60a generates a reduced area image for each reduction condition. Note that the region image may be an image of a predetermined region that occupies a part or all of the second image instead of the first image.

  The temporary comparison unit 60b compares the area image with each reduced area image based on significant pixels. In the present embodiment, the temporary comparison unit 60b calculates a numerical value corresponding to the ratio between the number of significant pixels in the region image and the number of significant pixels in the reduced region image. For example, the temporary comparison unit 60b calculates a ratio R of the number of significant pixels of the reduced area image to the number of significant pixels of the area image for each reduction condition. A significant pixel is a pixel having a pixel value “1” when the region image and each reduced region image are binarized (hereinafter referred to as an ON pixel).

  The selection unit 60c selects one reduction condition based on the comparison result by the temporary comparison unit 60b. In the present embodiment, the selection unit 60c selects a reduction condition when the ratio R becomes a value closest to the target value Rtg (for example, Rtg = 0.5) among the reduction conditions.

  The reduction unit 62 is realized with the central processing unit 40 as the center. The reduction unit 62 reduces the first image and the second image at the predetermined reduction rate. At this time, the reduction unit 62 reduces the first image and the second image under the reduction condition selected by the selection unit 60c. Thus, the reduction unit 62 generates a reduced first image and a reduced second image.

  The difference image generation unit 64 is realized with the central processing unit 40 as the center. The difference image generation unit 64 generates a difference image between the reduced first image and the reduced second image. In the present embodiment, a known feature point extraction process is executed on the binary image of the reduced first image to extract the feature points of the reduced first image. Similarly, feature point extraction processing is executed on the binary image of the reduced second image to extract feature points of the reduced second image. Then, the position of the corresponding feature point is compared between the reduced first image and the reduced second image, and a correction parameter for correcting the positional deviation between the reduced first image and the reduced second image is calculated. Here, a correction coefficient group for affine transformation of the reduced second image is calculated using the reduced first image as a reference. Then, after correcting the positional deviation between the reduced first image and the reduced second image according to the calculated correction coefficient group, a difference image between the reduced first image and the reduced second image is generated. That is, after the reduced second image is affine transformed according to the calculated correction coefficient group, a difference image between the reduced first image and the reduced second image is generated. Here, a difference image between the binary image of the reduced first image and the binary image of the reduced second image is generated. At this time, the differential image may be generated after expanding the ON pixels of the reduced first image or the reduced second image after the affine transformation.

  The integrated region determination unit 66 is realized with the central processing unit 40 as the center. The integrated region determination unit 66 determines a region (hereinafter referred to as an integrated region) that is a region in the difference image and satisfies a predetermined condition regarding significant pixel distribution. For example, the integrated region determination unit 66 determines the integrated region by performing any one of the following processes (1) to (4). Note that a significant pixel is an extracted pixel in the difference image (for example, a pixel whose absolute value is 1).

(1) The Euclidean distance between extracted pixels (for example, a pixel whose absolute value is 1) is obtained, and if the Euclidean distance is equal to or less than a predetermined threshold, the pixels are regarded as the same area and set as an integrated area.
(2) A rectangle having a predetermined size centered on the extracted pixel is created, and when the rectangles overlap each other, the pixels are regarded as the same region and set as an integrated region.
(3) A circumscribed rectangle is obtained for each connected pixel group of extracted pixels, and the circumscribed rectangle is expanded by a predetermined size. Then, when the extended circumscribed rectangles overlap, the connected pixel groups are regarded as the same region and set as an integrated region.
(4) A labeling process is performed on the reduced first image or the reduced second image after the affine transformation. Then, the label to which the extracted pixel belongs is determined, and the extracted pixel belonging to the same label is regarded as the same region and is defined as an integrated region.

  The corrected portion extraction unit 68 is realized with the central processing unit 40 as the center. The corrected portion extraction unit 68 compares the image of the corresponding region of the integrated region in the first image with the image of the corresponding region of the integrated region in the second image, and extracts the region of the corrected portion based on the comparison result. Hereinafter, the positional deviation correction unit 68a, the image division unit 68b, and the difference extraction unit 68c included in the corrected portion extraction unit 68 will be described.

  The positional deviation correction unit 68a corrects the positional deviation between the first image and the second image, and aligns the first image and the second image. In the present embodiment, the second image is affine transformed based on the reduction ratio and the correction coefficient group calculated by the difference image generation unit 64, whereby the first image and the second image are based on the first image. Perform alignment with.

  The image dividing unit 68b divides an image of the region corresponding to the integrated region (first divided image) from the first image, and divides an image of the region corresponding to the integrated region (second divided image) from the second image. For example, the image dividing unit 68b enlarges the difference image by the reciprocal of the predetermined reduction ratio, and calculates the coordinate range of the integrated area (that is, the corresponding area) enlarged by enlargement of the difference image. Then, the image of the calculated coordinate range (first divided image and second divided image) is divided from the first image and the second image that have been aligned.

  The difference extraction unit 68c compares the first divided image and the second divided image, and extracts a correction portion region based on the comparison result. Specifically, a difference between the binary image of the first divided image and the binary image of the second divided image is generated, and a pixel area extracted based on the difference is extracted as a corrected portion area.

  The corrected portion extraction unit 68 calculates the coordinate range of the integrated region in the reduced first image before alignment, enlarges the reduced first image to the original size, and expands the image of the enlarged coordinate range. (First divided image) may be divided from the first image. Similarly, after calculating the coordinate range of the integrated region in the reduced second image before alignment, the reduced second image is enlarged to the original size, and the image of the enlarged coordinate range (second divided image) May be divided from the second image. Then, the second divided image is affine transformed based on the reduction ratio and the correction coefficient group calculated by the difference image generation unit 64, whereby the first divided image and the second divided image are based on the first divided image. After performing the alignment with the image, a difference between the binary image of the first divided image and the binary image of the second divided image may be generated.

  Next, processing executed in the image processing apparatus 2 will be described with reference to FIGS. 4 and 5. 4 and 5 are flowcharts illustrating an example of processing executed in the image processing apparatus 2.

  The central processing unit 40 reduces the area image of the inspection area in the first image at a predetermined reduction rate while changing the reduction condition (S101). In this way, the central processing unit 40 generates a reduced area image for each reduction condition.

  Then, the central processing unit 40 calculates a ratio R of the number of ON pixels in the reduced area image to the number of ON pixels in the area image for each reduction condition (S102). Then, the central processing unit 40 selects a reduction condition when the ratio R becomes a value closest to the target value Rtg among the reduction conditions (S103).

  Then, the central processing unit 40 reduces the first image and the second image at a predetermined reduction rate under the reduction condition selected in step S103 (S104). In this way, the central processing unit 40 generates a reduced first image and a reduced second image.

  Then, the central processing unit 40 performs a known feature point extraction process on the binary image of the reduced first image, and extracts the feature points of the reduced first image. Similarly, feature point extraction processing is executed on the binary image of the reduced second image to extract feature points of the reduced second image (S105).

  Then, the central processing unit 40 compares the positions of corresponding feature points between the reduced first image and the reduced second image, and affine-transforms the reduced second image with reference to the reduced first image. The correction coefficient group is calculated (S106). Then, the central processing unit 40 corrects the positional deviation between the reduced first image and the reduced second image according to the calculated correction coefficient group (S107). That is, the reduced second image is affine transformed according to the calculated correction coefficient group.

  Then, the central processing unit 40 generates a difference image between the binary image of the reduced first image and the binary image of the reduced second image subjected to affine transformation (S108). At this time, the differential image may be generated after the ON pixels of the reduced first image or the reduced second image after the affine transformation are expanded by a predetermined size.

  Then, the central processing unit 40 determines an integrated area in the difference image (S109). For example, in the difference image, the central processing unit 40 creates a rectangle of a predetermined size centered on the extracted pixel, and when the rectangles overlap each other, the central processing unit 40 regards the pixels as the same region and sets it as an integrated region.

  Then, the central processing unit 40 corrects the positional deviation between the first image and the second image based on the predetermined reduction ratio and the correction coefficient group calculated in step S106 (S110). That is, the second image is affine transformed based on the predetermined reduction ratio and the correction coefficient group calculated in step S106.

  Then, the central processing unit 40 selects one integrated region (hereinafter referred to as a focused integrated region) from the integrated regions determined in step S109, and enlarges the difference image by the reciprocal of the predetermined reduction ratio. The coordinate range of the attention integration area is calculated (S111). Then, the central processing unit 40 divides the images (first divided image and second divided image) in the coordinate range calculated in step S111 from the first image and the second image in which the positional deviation is corrected, respectively. (S112). Then, a difference between the binary image of the first divided image and the binary image of the second divided image is generated, and the pixel area extracted based on the difference is extracted as the correction portion area (S113).

  The central processing unit 40 may divide the first divided image with a size larger than that of the second divided image. On this basis, the number of coincidence between the ON pixel of the binary image of the first divided image and the ON pixel of the binary image of the second divided image is calculated while moving the second divided image in the first divided image. You may make it produce | generate a difference in the state which moved the 2nd division image to the position in the 1st division image where the number of matches becomes the maximum.

  Then, the central processing unit 40 determines whether or not the steps of S111 to S113 have been executed for all the integrated areas determined in the step of S109 (S114). If there is an integrated area where the steps of S111 to S113 are not executed (N of S114), the central processing unit 40 executes the steps of S111 to S113 for the new target integrated area.

  On the other hand, when the steps S111 to S113 are executed for all the integrated regions (Y in S114), for example, a first image (composite image) in which the region of the corrected portion is filled with a predetermined color is generated and displayed on the display 48. It is displayed (S115).

  As described above, in the image processing apparatus 2, the first image and the reduced image (reduced first image) of the first image and the reduced image (reduced second image) of the second image are based on the difference image. A region (that is, a first divided image and a second divided image) for generating a difference from the second image is specified. Therefore, the processing load is reduced as compared with specifying the first divided image and the second divided image based on the difference between the first image and the second image. Therefore, in the image processing apparatus 2, the processing load when extracting the region of the corrected portion is relatively reduced.

  Further, the image processing apparatus 2 generates a difference image in a state where the positional deviation between the reduced first image and the reduced second image is corrected. Therefore, there is a high possibility that a region (that is, the first divided image and the second divided image) where a difference between the first image and the second image is to be generated is appropriately specified. As a result, it is possible to improve the extraction accuracy of the corrected portion region.

  Further, in the image processing apparatus 2, before actually reducing the first image and the second image, the image of the examination region that occupies a part or all of the first image or the second image (that is, the region image). Is reduced while changing the reduction condition, so that a reduction condition in which the ratio of ON pixels before and after reduction is closest to the target value is determined. Therefore, by appropriately setting the target value, even if the first image and the second image are reduced, the feature points of the first image and the second image are as small as possible in the reduced first image and the reduced second image. Can be maintained. Therefore, the accuracy of correcting the positional deviation between the reduced first image and the reduced second image is improved, and as a result, the extraction accuracy of the corrected portion is improved. As a result, the accuracy of correcting the positional deviation between the first image and the second image is also improved. From this point as well, the extraction accuracy of the corrected portion is improved.

  The present invention is not limited to the above embodiment.

  For example, in the above embodiment, the ratio of the number of ON pixels before and after the reduction of the area image is calculated under the reduction condition in which the feature points of the first image and the second image are maintained as much as possible in the reduced first image and the reduced second image. However, the reduction condition may be specified by another method. Two other methods for specifying the reduction condition will be described below.

  For example, in the first method, the central processing unit 40 executes the feature point extraction processing on the binary image of the region image and the binary image of the reduced region image in step S102, thereby reducing the region image and the reduced image. After extracting the feature points of the region image, the number of feature points that are also present in the reduced region image among the feature points of the region image is counted for each reduction condition. In step S103, a reduction condition is selected when the number counted in step S102 is the largest.

  Further, for example, in the second method, the central processing unit 40 enlarges the reduced area image to the size before reduction in step S102, and the binary image of the enlarged reduced area image and the binary image of the area image. An inter-pixel calculation is performed between the images and a numerical value corresponding to the degree of coincidence between the two binary images is calculated. For example, the central processing unit 40 calculates an exclusive OR between pixels corresponding to positions, and determines the number of pixels for which the exclusive OR is “true” (that is, only one pixel is an ON pixel). Count. Then, the central processing unit 40 selects a reduction condition when the counted number becomes the minimum in step S103. Further, for example, the central processing unit 40 calculates an exclusive OR and a logical product between pixels corresponding to positions, and a pixel in which the logical product is “true” (that is, both pixels are ON pixels). The ratio of the number of pixels for which the exclusive OR is “true” is calculated. Then, the central processing unit 40 selects a reduction condition when the calculated ratio is minimized in step S103. Further, for example, the central processing unit 40 calculates a logical sum and a logical product between pixels corresponding to positions, and the number of pixels in which the logical product is “true” and the logical sum is “true” (that is, at least The number of pixels in which one pixel is an ON pixel) is counted. A difference obtained by subtracting the number of pixels whose logical sum is “true” from the number of ON pixels in the region image; a difference obtained by subtracting the number of pixels where the logical product is “true” from the number of ON pixels in the region image; The product of is calculated. Then, the central processing unit 40 selects a reduction condition when the calculated product is minimized in step S103. Further, for example, the central processing unit 40 calculates a logical sum and logical product between pixels corresponding to positions, the number of pixels for which the logical product is “true”, the number of pixels for which the logical sum is obtained, The sum of is calculated. Then, the central processing unit 40 selects a reduction condition when the calculated sum becomes maximum in step S103.

  In the above embodiment, the image reduction method performed in step S104 is selected from a plurality of image reduction methods. However, the image reduction method performed in step S104 may be a predetermined image reduction method. . For example, the image reduction method performed in step S104 may be OR reduction. Hereinafter, an aspect in which the first image and the second image are OR-reduced at a predetermined reduction rate in step S104 will be described.

  In this aspect, the central processing unit 40 binarizes the first image and the second image in advance, and OR reduces the region image at a predetermined reduction rate while changing the threshold pixel number as a reduction condition in step S101. . In step S104, the first image and the second image are OR-reduced at the predetermined reduction rate under the reduction condition (that is, the threshold pixel number) selected in step S103. In this way, even when the first image and the second image are OR-reduced to extract the correction portion area, the processing load is reduced and the extraction accuracy of the correction portion area is improved.

It is a figure showing an example of the hardware constitutions of the image processing apparatus which concerns on embodiment of this invention. It is a figure which shows a mode that the area | region of a correction part is extracted based on a 1st image and a 2nd image. It is a block diagram of the function implement | achieved with the image processing apparatus which concerns on embodiment of this invention. It is a flowchart figure which shows an example of the process performed with the image processing apparatus concerning embodiment of this invention. It is a flowchart figure which shows an example of the process performed with the image processing apparatus concerning embodiment of this invention.

Explanation of symbols

  2 image processing apparatus, 40 central processing unit, 42 main storage unit, 44 auxiliary storage unit, 46 operation unit, 48 display, 50 data bus, 60 reduction condition determination unit, 60a temporary reduction unit, 60b temporary comparison unit, 60c selection unit 62 reduction part, 64 difference image generation part, 66 integrated area determination part, 68 correction part extraction part, 68a position shift correction part, 68b image division part, 68c difference extraction part.

Claims (11)

Reduction means for reducing the first image scanned from the original and the second image scanned from the corrected original corrected on the original at a predetermined reduction rate;
Difference image generation means for generating a difference image between the reduced first image and the reduced second image;
A region determination means for determining a region in the difference image that satisfies a predetermined condition regarding the distribution of significant pixels;
The image of the corresponding area of the determined area in the first image is compared with the image of the corresponding area of the determined area in the second image, and the correction area is extracted based on the comparison result. Correction portion extraction means;
An image processing apparatus comprising: Feature extraction means for extracting feature points of the reduced first image and the reduced second image based on significant pixels;
Calculation means for calculating a correction parameter for correcting a positional deviation between the reduced first image and the reduced second image based on the extracted feature points;
The difference image generation means generates the difference image after correcting a positional shift between the reduced first image and the reduced second image according to the correction parameter;
The image processing apparatus according to claim 1. A temporary reduction means for reducing an area image of an inspection area occupying a part or all of the first image or the second image at the predetermined reduction rate while changing a reduction condition;
Comparison means for comparing the region image with the reduced region image based on significant pixels;
Selecting means for selecting one reduction condition based on the result of the comparison by the comparison means;
The reduction means reduces the first image and the second image under the selected reduction condition;
The image processing apparatus according to claim 2. The comparing means calculates a numerical value corresponding to a ratio between the number of significant pixels of the region image and the number of significant pixels of the reduced region image;
The selection means selects a reduction condition when the numerical value is closest to a target value;
The image processing apparatus according to claim 3. The comparing means extracts the feature points of the region image and the reduced region image based on significant pixels, and calculates the number of feature points that are also present in the reduced region image among the feature points of the region image. Means for counting, and the selecting means selects a reduction condition when the number is the largest,
The image processing apparatus according to claim 3. A temporary reduction means for reducing an area image of an inspection area occupying a part or all of the first image or the second image at the predetermined reduction rate while changing a reduction condition;
Enlarging means for enlarging the reduced area image to the size before reduction,
Comparison means for comparing the enlarged reduced area image with the area image based on significant pixels;
Selecting means for selecting one reduction condition based on the result of the comparison by the comparison means;
The reduction means reduces the first image and the second image under the selected reduction condition;
The image processing apparatus according to claim 2. The comparison means calculates a numerical value corresponding to the degree of coincidence between the enlarged reduced area image and the area image by performing a logical operation between pixels corresponding to positions,
The selection means selects a reduction condition when the numerical value reaches a peak value;
The image processing apparatus according to claim 6. The correction portion extraction means aligns the image of the corresponding area in the first image and the image of the corresponding area in the second image based on the correction parameter, and then performs the first adjustment. Comparing the image of the corresponding area in the image with the image of the corresponding area in the second image;
The image processing apparatus according to claim 2, wherein the image processing apparatus is an image processing apparatus. The reduction condition is an image reduction method,
An image processing apparatus according to claim 3, wherein: The reduction means OR-reduces the first image and the second image,
The temporary reduction means OR reduces the area image,
The reduction condition is a threshold pixel number for OR reduction,
An image processing apparatus according to claim 3, wherein: Reduction means for reducing the first image scanned from the original and the second image scanned from the corrected original corrected on the original at a predetermined reduction ratio;
Difference image generation means for generating a difference image between the reduced first image and the reduced second image;
An area determination means for determining an area in the difference image that satisfies a predetermined condition regarding a significant pixel distribution;
The image of the corresponding area of the determined area in the first image is compared with the image of the corresponding area of the determined area in the second image, and the correction area is extracted based on the comparison result. Correction part extraction means,
As a program that allows the computer to function.

Images