JP2009290674A - Original collating apparatus, original collating method and original collate program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To accurately implement collation with an original, regardless of direction, front side and rear side of a document when reading the document. <P>SOLUTION: An original collating apparatus 1 includes: a reading section 11 for reading image information of a document; an extraction section 18a for extracting information about a plurality of preset images to be determined from image information of the document read by the reading section 11; a coordinate transforming section 15 for performing coordinate transform on information of a reference image of the original registered beforehand or on the information about the plurality of images to be determined extracted by the extraction section 18a; and a paper-print detection section 14 for comparing the information about the plurality of images to be determined with the information about the reference image of the original registered beforehand for the information coordinate-transformed by the coordinate transforming section 15, so as to determine whether the document is original or not. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to an original collation apparatus, an original collation method, and an original collation program for collating whether or not a read original is an original.

  Patent Document 1 discloses a technique for preventing a copied material obtained by copying banknotes or securities from a copying machine or a printer from being misused. In this technique, in order to collate the read original with a reference image (for example, 32 × 32 pixels) at a specific position of the original, for example, 64 × 64 pixels of the read image corresponding to the position of the reference image are collated. As an image, the correlation value by the normalized correlation method is calculated while shifting one pixel at a time in the X and Y directions.

JP 2005-38389 A

  In a technique for performing authenticity determination by comparing image information of a read original and information of an original reference image, it is necessary to compare the original image at the same position as or corresponding to the position of the original reference image. At this time, there arises a problem that the original reference image position and orientation cannot be matched with each other due to a difference in reading direction and front and back of the document, and erroneous determination is made.

  SUMMARY OF THE INVENTION An object of the present invention is to provide a technique capable of accurately verifying an original regardless of the reading direction and the front and back of a document to be determined.

  The invention according to claim 1 of the present application includes reading means for reading image information of a document, extraction means for extracting information on a plurality of determination target images set in advance from image information of the document read by the reading means, Coordinate conversion means for converting the information of the registered reference image of the original or the information of the plurality of determination target images extracted by the extraction means, and the information of the plurality of determination target images regarding the information subjected to coordinate conversion by the coordinate conversion means An original collating apparatus having a determination unit that compares information with information of a reference image of an original registered in advance and determines whether or not the original is an original.

  Further, in the invention according to claim 2 of the present application, the reading unit reads image information on the front and back sides of the document, and the extraction unit reads each of the plurality of determination targets from the image information on the front and back sides of the document read by the reading unit. 2. The original collating apparatus according to claim 1, wherein an image is extracted, and the determination unit compares the information of the plurality of determination target images on the front and back of the document extracted by the extraction unit with the information of the reference image. is there.

  The invention according to claim 3 of the present application further includes a side detection unit that detects the position of a side of the document from the image information of the document read by the reading unit, and the extraction unit is detected by the side detection unit. 2. The original collating apparatus according to claim 1, wherein information of the plurality of determination target images is extracted with reference to a position of a corner constituted by two adjacent sides of the original.

  Further, the invention according to claim 4 of the present application is provided with notifying means for notifying the image information on the opposite side of the original to be read by the reading means when the determining means determines that the original is not the original. 1 is an original collating apparatus according to 1;

  The invention according to claim 5 of the present application is the original collating apparatus according to claim 1, wherein the extraction position of the information of the plurality of determination target images is a position with reference to each corner of the document.

  The invention according to claim 6 includes a step of reading image information of a document, a step of extracting information of a plurality of determination target images set in advance from the read image information of the document, and a pre-registered original A step of performing coordinate conversion on the reference image information or information on the plurality of determination target images extracted by the extracting unit; and on the coordinate-converted information, the information on the plurality of determination target images and the previously registered original reference image A method of comparing originals, and a step of determining whether or not the original is an original.

  The invention according to claim 7 includes a step of capturing image information of a document, a step of extracting information of a plurality of determination target images set in advance from the captured image information of the document, and a pre-registered original A step of performing coordinate conversion on the reference image information or information on the plurality of determination target images extracted by the extraction unit; and on the coordinate-converted information, information on the plurality of determination target images and a pre-registered original reference image An original collating program that causes a computer to execute a step of comparing with information and determining whether or not the original is an original.

  According to the present invention, it is possible to accurately collate with the original regardless of the orientation and front / back of the document.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings.

<Original verification device>
The original collation apparatus according to the present embodiment collates an original from which image information is to be read with the original, and information on a determination target image of image information of the read original and an original reference image registered in advance. The degree of coincidence with this information is calculated to determine whether or not the document is an original. The technique disclosed in Patent Document 1 is used for the specific image coincidence calculation applied in the original collation apparatus of the present embodiment. Such an original collating apparatus according to the present embodiment is applied to a device provided with means for reading image information of a document. For example, a scanner, a copier, or a multifunction machine.

<First Embodiment: Device Configuration>
FIG. 1 is a configuration diagram illustrating an original collating apparatus according to the first embodiment. The original collation apparatus 1 according to the first embodiment includes a reading unit 11, an AD conversion unit 12, an image processing unit 13, a paper pattern detection unit 14, a coordinate conversion unit 15, a timing generation unit 16, an operation panel 17, a control unit 18, and A system controller 19 is provided.

  The reading unit 11 reads an image of a document and generates image information. For example, an image input portion of a scanner, an image input portion of a copier, or a multifunction peripheral.

  The AD conversion unit 12 is a part that performs processing for converting analog image information read by the reading unit 11 into digital.

  The image processing unit 13 is a part that inputs the image information converted into a digital signal by the AD conversion unit 12 and performs predetermined image processing. For example, the white reference is adjusted and the color tone such as gamma correction is adjusted.

  The paper pattern detection unit 14 uses information on the pattern (paper pattern) formed by the entanglement of the fibrous material of the paper from the image information, and matches the information on the read paper pattern and the original paper pattern information. This is a determination part that calculates the degree and compares the read original with the original. Specifically, as disclosed in Patent Document 1, a determination target image at a position corresponding to a reference image at a specific position of the original is extracted from the image information of the read original, and the reference image information and the determination target image are extracted. The correlation value by the normalized correlation method is calculated while shifting the above information in the X and Y directions one pixel at a time. Then, it is determined whether or not the read original is an original by comparing the correlation value or the value of the normalized score obtained from the correlation value with a predetermined threshold value.

  The coordinate conversion unit 15 is a part that performs coordinate conversion on a predetermined determination target image in the information on the original reference image registered in advance or the image information on the read document. The coordinate conversion unit 15 includes an address conversion unit 15a and a rotation conversion unit 15b. The address conversion unit 15a performs coordinate conversion of image information in the X and Y directions, and the rotation conversion unit 15b performs coordinate conversion related to rotation of the image information.

  The timing generation unit 16 is a part that provides a timing signal for synchronizing processing to the reading unit 11 and the AD conversion unit 12.

  The operation panel 17 is, for example, a touch panel display that serves as both display means and input means. The display means of the operation panel 17 also functions as a notification unit. When it is determined that the read original is an original, a notification to that effect is given. When it is determined that the original is not the original, the opposite side of the original is read. A notice to that effect and a notice to the effect that it is not the original.

  The control unit 18 mainly gives an instruction to the image processing unit 13 and the coordinate conversion unit 15 and receives a determination result sent from the paper pattern detection unit 14 to perform a predetermined process. In particular, in the original collating apparatus 1 of the present embodiment, the control unit 18 implements an extraction unit 18a that extracts information on a plurality of determination target images set in advance from image information of a document read by the reading unit 11. .

  The extraction unit 18a of the control unit 18 extracts a determination target image at a position based on the corner of the read document. Here, extracting a plurality of predetermined determination target images from the image information of the read original is a function of extracting an original image area and orientation for comparing the reference image and the determination target image of the original. In other words, a plurality of determination target images are set according to how the document is placed. Specifically, according to the timing generated by the timing generation unit 16, an instruction to extract image information having predetermined XY coordinates is given to the image processing unit 13. As a result, the extracted image information is sent from the image processing unit 13 to the paper pattern detection unit 14.

  The system controller 19 is a part that controls each part, and controls the entire apparatus.

  Further, the original collating apparatus 1 according to the present embodiment includes an automatic document feeder 21 and a document size detecting unit 22 as necessary. The automatic document feeder 21 has a function of automatically feeding a document to be read to the reading unit 11. The document size detection unit 22 has a function of detecting the size of the document placed on the reading unit 11 and determining the orientation of the document.

  In the original collating apparatus 1 of the present embodiment having such a configuration, predetermined coordinate conversion is performed on the information of the reference image of the original registered in advance or the information of the plurality of determination target images extracted by the extracting unit 18a of the control unit 18. This is performed by the coordinate conversion unit 15. Further, the paper pattern detection unit 14 compares the information of the plurality of determination target images with the information of the reference image of the original registered in advance for the information subjected to the coordinate conversion by the coordinate conversion unit 15, and the read original is the original. It is determined whether or not there is.

<First Embodiment: Original Verification Method>
First, in accordance with an instruction on the operation panel 17, the system controller 19 reads information on the original reference image registered in the storage unit M in advance and transfers it to the control unit 18.

  Next, the reading unit 11 reads image information of a document to be collated. In addition to being placed directly at the reading position of the reading unit 11, the document may be sent to the reading position by the automatic document feeder 21.

  Next, the image information of the document read by the reading unit 11 is converted into a digital signal by the AD conversion unit 12, and color adjustment such as white reference adjustment and gamma correction is performed by the image processing unit 13. Thereafter, information of a plurality of determination target images set in advance is extracted from the image information of the document by the extraction unit 18a of the control unit 18.

  Here, the positions of the plurality of determination target images extracted by the control unit 18 are positions based on the corners of the read document. For example, when a rectangular document is read, determination target images at four positions with respect to each corner of the rectangle are extracted. Further, when a triangular document is read, determination target images at three positions with respect to each corner of the triangle are extracted. This is based on the premise that the document is placed on the reading unit 11 with a corner as a reference, regardless of which corner the document is placed on, the determination target corresponding to the position of the original reference image This is so that an image can be obtained.

  Next, the coordinate conversion unit 15 uses the reference position of the reference corner of the original reference image information registered in the storage unit M or the information of the plurality of determination target images extracted by the extraction unit 18 a of the control unit 18. Only the coordinates in the X and Y directions according to the coordinates are converted by the address converter 15a, and the coordinates are converted by the rotation converter 15b so as to rotate by an angle corresponding to the position of the reference corner.

  Next, the paper pattern detection unit 14 collates the information subjected to coordinate conversion by the coordinate conversion unit 15 with a plurality of determination target images and the original reference image. In other words, using the information on the determination target image or the reference image that has undergone coordinate transformation, all of the plurality of determination target images are checked against the original reference image, and at least one is determined to be the original If there is, it is determined that the read original is the original. If it is determined to be the original, it is notified to the operation panel 17 via the system controller 19.

  On the other hand, if all of the plurality of determination target images are not determined to be originals, it is determined that the read original is not the original, or the operation panel 17 is notified that the opposite side of the original is read. This is for avoiding this by reading the opposite side of the original again because the read side of the original may be opposite to the front and back when the original reference image is registered.

  At this time, since the user may make a mistake in reversing the document, a part of the image information at the time of collation of the previously captured document surface is stored, and the document surface temporarily stored when the image information on the back side of the document is captured It may be confirmed that the image information does not match the image information. If the automatic document feeder 21 is provided, the reading determination mode on the platen may be prohibited, and the automatic document feeder 21 may automatically reverse the document.

  Even when the opposite side of the document is scanned, multiple images to be judged are extracted in the same way as before, all of them are checked against the original reference image, and if any of them is judged to be the original, it is scanned. The original is determined to be the original. If it is determined to be the original, it is notified to the operation panel 17 via the system controller 19.

  On the other hand, if all of the plurality of determination target images are not determined to be originals, it is determined that the read original is not an original. Here, since both sides of the document have already been collated, notification that the opposite side of the document is read again is not performed, and it is determined that the document is not the original document. When it is determined that it is not the original, it is notified to the operation panel 17 via the system controller 19.

<Specific example>
FIG. 2 is a schematic diagram illustrating an original reference image. The original reference image is extracted from the original image information by a predetermined number of pixels around the coordinates (X0, Y0) in the X and Y directions with reference to a predetermined origin. The original reference image is information uniquely determined for the original paper, for example, information on a pattern (paper pattern) formed by the entanglement of the fibrous material on the original paper. The reference image is registered as image information of 32 × 32 pixels, for example.

  In order to collate the read original image with the original reference image, it is necessary to extract a determination target image corresponding to the position of the original reference image from the read original image information. That is, the determination target image is extracted in an area larger than the reference image with the X and Y coordinates of the read image being the same as the center coordinates (X0, Y0) of the original reference image registered in advance.

  At this time, if the placement state of the original to be read is different from the placement state when the original is registered, the origin is shifted, so that the determination target image having the same X and Y coordinates as the reference image is extracted. However, the determination target image at a position different from the position of the reference image is extracted.

  Therefore, in the present embodiment, a plurality of determination target images are extracted in consideration of the document placement state. FIG. 3 is a schematic diagram for explaining the relationship between the document placement state and the determination target image. In the placement state shown in FIG. 3A, the document is placed with the corner a of the original as the origin, and the same case as the original placement state when the original reference image is registered is assumed. . In this case, the determination target image is extracted around the same coordinates (Xa, Ya) as the center coordinates (X0, Y0) of the original reference image.

  The placement state shown in FIG. 3B is a state in which the corner b of the document is placed as the origin, and is rotated 90 degrees to the right with respect to the placement state of the original when the original reference image is registered. Assumes that. In this case, the determination target image is extracted with the coordinates (Xb, Yb) obtained by rotating the original 90 degrees to the right with respect to the center coordinates (X0, Y0) of the original reference image.

  The placement state shown in FIG. 3C is a state where the document is placed with the corner c of the document as the origin, and is rotated 180 degrees with respect to the placement state of the original when the original reference image is registered. Is assumed. In this case, the determination target image is extracted around the coordinates (Xc, Yc) obtained by rotating the document by 180 degrees with respect to the center coordinates (X0, Y0) of the original reference image.

  The placement state shown in FIG. 3D is a state where the document is placed with the corner d of the document as the origin, and is rotated 90 degrees to the left with respect to the placement state of the original when the original reference image is registered. Assumes that. In this case, the determination target image is extracted with the coordinates (Xd, Yd) obtained by rotating the document 90 degrees to the left with respect to the center coordinates (X0, Y0) of the original reference image.

  FIG. 4 is a schematic diagram illustrating the position of the determination target image extracted in the present embodiment. As described above, in order to be able to extract the determination target image corresponding to the reference image regardless of what even portion of the document is placed as a reference, FIG. 4A shows the image information of the read document. As shown, a determination target image centered on four locations of coordinates (Xa, Ya), coordinates (Xb, Yb), coordinates (Xc, Yc), and coordinates (Xd, Yd) is extracted.

  Note that in the apparatus provided with the document size detection unit 22, when the orientation of the document is known from the detection of the size of the document to be read, FIGS. 4 (b) and 4 (c). As shown, it is only necessary to extract two determination target images in accordance with the orientation of the corresponding document. For example, when the document is placed vertically, as shown in FIG. 4B, two positions of coordinates (Xa, Ya) and coordinates (Xc, Yc) corresponding to the portrait placement are set. A determination target image centered is extracted.

  On the other hand, when the document is placed horizontally, as shown in FIG. 4 (c), two positions of coordinates (Xb, Yb) and coordinates (Xd, Yd) corresponding to the horizontally placed document are displayed. A determination target image centered is extracted.

  After extracting the plurality of determination target images, the coordinate conversion unit 15 performs address conversion and rotation conversion on the reference image information or the plurality of determination target image information according to the reference corner position.

  In this specific example, the case where the address conversion and the rotation conversion are performed on the information of the reference image will be described. One of the information may be subjected to address conversion, and the other may be subjected to rotation conversion.

  For example, in the determination target image centered on the coordinates (Xa, Ya), the amount of address conversion and rotation conversion is zero because it is the same as the original placement state when the original reference image is registered.

  In addition, in the determination target image centered on the coordinates (Xb, Yb), the original reference image is rotated 90 degrees to the right with respect to the original placement state when the original reference image was registered. The X and Y coordinates are converted by the address conversion unit so that the center coordinates (X0, Y0) become the coordinates (Xb, Yb), and the rotation conversion unit performs a rotation that rotates 90 degrees to the right.

  In addition, in the determination target image centered on the coordinates (Xc, Yc), the center coordinates of the original reference image are rotated by 180 degrees with respect to the original placement state when the original reference image is registered. The address converter converts the X and Y coordinates so that (X0, Y0) becomes the coordinates (Xc, Yc), and the rotation converter converts 180 degrees.

  In addition, the determination target image centered on the coordinates (Xd, Yd) is rotated 90 degrees to the left with respect to the original placement state when the original reference image was registered. The address conversion unit converts the X and Y coordinates so that the center coordinate (X0, Y0) becomes the coordinate (Xd, Yd), and the rotation conversion unit performs a rotation that rotates 90 degrees to the left.

  The paper pattern detection unit 14 collates the information of the reference image thus coordinate-converted with all four determination target images. If any one of these is determined to be an original, it is determined that the read original is the original, and this is notified to the operation panel.

  Further, when it is determined that all four determination target images are not the original as a result of collation with the reference image information, the operation panel 17 is notified to read the opposite side of the document. In consideration of the fact that the front and back sides of the original are reversed and placed, the opposite side is read in the same manner, and the same plurality of determination target images as described above are extracted and coordinate-converted reference image information is used. This is for collation. If it is determined in this collation that at least one of the plurality of determination target images is an original, the read original is determined to be an original, and a notification to that effect is sent to the operation panel.

  On the other hand, if it is determined that all four determination target images are not originals as a result of collation with the information of the reference image, it is determined that the read original is not the original, and the operation panel 17 is notified accordingly.

<Second Embodiment: Device Configuration>
FIG. 5 is a configuration diagram illustrating an original collating apparatus according to the second embodiment. The original collating apparatus 1 according to the second embodiment includes a front reading unit 11a, a back reading unit 11b, a front AD conversion unit 12a, a back AD conversion unit 12b, a front image processing unit 13a, a back image processing unit 13b, and a paper pattern detection unit. 14, a coordinate conversion unit 15, a table timing generation unit 16 a, an image memory 23, an operation panel 17, a control unit 18, and a system controller 19.

  The front reading unit 11a reads an image on the front side (one side) of the document and generates image information. The back reading unit 11b reads the image on the back side (the other side) of the document and generates image information. These are, for example, an image input part of a scanner, and an image input part of a copying machine or a multifunction machine.

  The table AD conversion unit 12a is a part that performs processing for converting analog image information read by the table reading unit 11a into digital. The back AD converter 12b is a part that performs processing for converting analog image information read by the back reader 11b into digital.

  The front image processing unit 13a is a part that inputs the image information converted into a digital signal by the front AD conversion unit 12a and performs predetermined image processing. The back image processing unit 13b is a part that inputs image information converted into a digital signal by the back AD conversion unit 12b and performs predetermined image processing. These adjust, for example, a white reference or a color tone such as gamma correction.

  The paper pattern detection unit 14 uses information on the pattern (paper pattern) formed by the entanglement of the fibrous material of the paper from the image information, and matches the information on the read paper pattern and the original paper pattern information. This is a determination part that calculates the degree and compares the read original with the original. Specifically, as disclosed in Patent Document 1, a determination target image at a position corresponding to a reference image at a specific position of the original is extracted from the image information of the read original, and the reference image information and the determination target image are extracted. The correlation value by the normalized correlation method is calculated while shifting the above information in the X and Y directions one pixel at a time. Then, it is determined whether or not the read original is an original by comparing the correlation value or the value of the normalized score obtained from the correlation value with a predetermined threshold value.

  The coordinate conversion unit 15 is a part that performs coordinate conversion on a predetermined determination target image in the information on the original reference image registered in advance or the image information on the read document. The coordinate conversion unit 15 includes an address conversion unit 15a and a rotation conversion unit 15b. The address conversion unit 15a performs coordinate conversion of image information in the X and Y directions, and the rotation conversion unit 15b performs coordinate conversion related to rotation of the image information.

  The table timing generation unit 16a is a part that provides a timing signal for synchronizing processing to the table reading unit 11a and the table AD conversion unit 12a. The back timing generation unit 16b is a part that provides a timing signal for synchronizing processing to the back reading unit 11b and the back AD conversion unit 12b.

  The operation panel 17 is, for example, a touch panel display that serves as both display means and input means. The display means of the operation panel 17 also functions as a notification unit. When it is determined that the read original is an original, a notification to that effect is given. When it is determined that the original is not the original, the opposite side of the original is read. A notice to that effect and a notice to the effect that it is not the original.

  The control unit 18 mainly gives instructions to the front image processing unit 13a, the back image processing unit 13b, and the coordinate conversion unit 15, and receives a determination result sent from the paper pattern detection unit 14 and performs predetermined processing. . In particular, in the original collating apparatus 1 of the present embodiment, the control unit 18 extracts information on a plurality of determination target images set in advance from the image information on each of the front and back sides of the document read by the front reading unit 11a and the back reading unit 11b. The extraction unit 18a is realized.

  The extraction unit 18a of the control unit 18 extracts a determination target image at a position based on the corner of the read document. Specifically, for the image information on the surface of the document, an instruction to extract image information having predetermined XY coordinates is given to the table image processing unit 13a in accordance with the timing generated by the table timing generation unit 16a. As a result, the extracted image information is sent from the front image processing unit 13 a to the paper pattern detection unit 14. For the image information on the back side of the document, an instruction to extract image information having predetermined XY coordinates is given to the back image processing unit 13b according to the timing generated by the back timing generation unit 16b. As a result, the extracted image information is sent from the back image processing unit 13b to the paper pattern detection unit 14.

  In the present embodiment, the coordinate conversion unit 15 is provided in the control unit 18, the coordinate conversion of the reference image is performed in advance, a plurality of reference images corresponding to the orientation of the document are generated, held in the image memory, and the plurality of reference images are stored. Comparison with the determination target image is performed. There are a maximum of four types of coordinate-converted reference images including the original image (0 ° original, 90 °, 180 °, 270 °), and the short / long side in the document reading direction can be determined. Two types of reference images that have undergone coordinate conversion are required, and four types of reference images that have undergone coordinate conversion are required when the short / long side is unknown. The coordinate-converted reference image can be used for both front and back.

  The system controller 19 is a part that controls each part, and controls the entire apparatus.

  Further, the original collating apparatus 1 according to the present embodiment includes an automatic document feeder 21 and a document size detecting unit 22 as necessary. The automatic document feeder 21 has a function of automatically feeding a document to be read to the reading unit 11. The document size detection unit 22 has a function of detecting the size of the document placed on the reading unit 11 and determining the orientation of the document.

  In the original collating apparatus 1 of this embodiment having such a configuration, the original to be determined is read simultaneously, and the front and back can be collated without inverting the original. It can be performed. As another method, the back side image information is stored in the image memory 23, the back side image information is read from the image memory 23 after the front side collation, and the back side image paper crest is collated. The detection unit can be shared on both sides.

<Second Embodiment: Original Verification Method>
First, in accordance with an instruction on the operation panel 17, the system controller 19 reads information on the original reference image registered in the storage unit M in advance and transfers it to the control unit 18.

  Next, the front reading unit 11a reads the image information on the front side of the document to be collated, and the back reading unit 11b reads the image information on the back side of the document to be collated. In addition to being placed at the reading position of the front direct reading unit 11a and the back reading unit 11b, the document may be sent to the reading position by the automatic document feeder 21. In the case where the front direct reading unit 11a and the reverse reading unit 11b are incorporated in the automatic document feeder 21, the front direct reading unit 11a and the reverse reading are performed while the original is being fed by the automatic document feeder. It is read by the unit 11b.

  Next, image information on the surface of the original read by the table reading unit 11a is converted into a digital signal by the table AD conversion unit 12a, and color adjustment such as white reference adjustment and gamma correction is performed by the table image processing unit 13a. In addition, image information on the back side of the document read by the back reading unit 11b is converted into a digital signal by the back AD conversion unit 12b, and color adjustment such as white reference adjustment and gamma correction is performed by the back image processing unit 13b.

  Thereafter, the information of a plurality of determination target images set in advance is extracted from the image information on the front and back sides of the document by the extraction unit 18a of the control unit 18. Information on the plurality of determination target images on the surface of the document extracted by the extraction unit 18a is sent to the paper pattern detection unit 14a. On the other hand, information on a plurality of determination target images on the back side of the document extracted by the extraction unit 18a is also sent to the paper pattern detection unit 14b.

  Here, the positions of the plurality of determination target images extracted by the control unit 18 are the front and back positions on the basis of the corner of the read document. For example, when a rectangular document is read, determination target images in a total of eight positions are extracted, with four positions on the front and back with reference to each corner of the rectangle. Further, when a triangular document is read, determination target images are extracted at a total of six positions, three on the front and back with reference to each corner of the triangle. This is based on the premise that the document is placed on the reading unit 11 with a corner as a reference, regardless of which corner the document is placed on, the determination target corresponding to the position of the original reference image This is so that an image can be obtained.

  Next, the coordinate conversion unit 15 serves as a reference for information on the original reference image registered in the storage unit M in advance or information on a plurality of determination target images on the surface of the document extracted by the extraction unit 18a of the control unit 18. Only the coordinates in the X and Y directions corresponding to the corner positions are converted by the address conversion section 15a, and the coordinates are converted by the rotation conversion section 15b so as to rotate by the angle corresponding to the reference corner position, and the image memory is used. 23 temporarily stores a plurality of reference images.

  Next, the paper pattern detection unit 14 collates a plurality of reference images stored in the image memory 23 coordinate-converted by the coordinate conversion unit 15 with a plurality of determination target images. That is, using the information of the determination target image subjected to coordinate conversion or the information of the reference image subjected to coordinate conversion, all of the plurality of determination target images are compared with the original reference image, and one of the front and back is the original. If it is determined that the original is read, it is determined that the original is the original. If it is determined to be the original, it is notified to the operation panel 17 via the system controller 19.

  On the other hand, if all of the plurality of determination target images on the front and back sides of the document are not determined to be originals, it is determined that the read original is not the original. When it is determined that it is not the original, it is notified to the operation panel 17 via the system controller 19.

<Specific example>
As a specific example of the original collating apparatus 1 in the second embodiment, the extraction of a plurality of determination target images for the original reference image and the surface of the original is the same as in the first embodiment shown in FIGS.

  FIG. 6 is a schematic diagram for explaining the relationship between the document placement state and the determination target image on the back surface. In the placement state shown in FIG. 6 (a), it is assumed that the document is placed with the corner d of the original as the origin, and the case where the original reference image of the original is reversed from the original placement state when the original reference image is registered is assumed. ing. In this case, the determination target image is extracted with the coordinates (Xd ′, Yd ′) at the same position as the center of the reference image when the original is turned back.

  The placement state shown in FIG. 6B is a state in which the corner c of the document is placed as the origin, which is reversed from the original placement state when the original reference image was registered, It is assumed that the angle is rotated 90 degrees. In this case, the determination target image is extracted with the coordinates (Xc ′, Yc ′) obtained by rotating the original 90 degrees to the right with respect to the center of the reference image when the original is turned back.

  The placement state shown in FIG. 6C is a state where the original is placed with the corner b of the original as the origin, and is reversed with respect to the original placement state when the original reference image is registered. It is assumed that the angle is rotated. In this case, the determination target image is extracted around the coordinates (Xb ′, Yb ′) obtained by rotating the document by 180 degrees with respect to the center of the reference image when the original is turned back.

  The placement state shown in FIG. 6D is a state where the original is placed with the corner a of the original as the origin, and the left and right sides are reversed with respect to the original placement state when the original reference image is registered. It is assumed that the angle is rotated 90 degrees. In this case, the determination target image is extracted with the coordinates (Xa ′, Ya ′) obtained by rotating the document 90 degrees to the left with respect to the center of the reference image when the original is turned back.

  FIG. 7 is a schematic diagram illustrating the position of the determination target image extracted in the present embodiment. As described above, in order to be able to extract the determination target image corresponding to the reference image regardless of what even portion of the document is placed as a reference, FIG. 7A shows the image information of the read document. As shown in the figure, for the surface of the document, a determination target image centered on the coordinates (Xa, Ya), coordinates (Xb, Yb), coordinates (Xc, Yc), and coordinates (Xd, Yd) is extracted. With respect to the back side of the document, the determination target is centered on four points of coordinates (Xa ′, Ya ′), coordinates (Xb ′, Yb ′), coordinates (Xc ′, Yc ′), and coordinates (Xd ′, Yd ′). Extract images.

  If the orientation of the original is known from the detection of the size of the original to be read in the apparatus provided with the original size detection unit 22, FIG. 7B and FIG. As shown in the drawing, it is only necessary to extract the determination target images at two positions on the front and back according to the orientation of the corresponding document.

  For example, when the document is placed vertically, as shown in FIG. 7B, the coordinates (Xa, Ya) and the coordinates (Xc, Yc) are on the surface of the document corresponding to the portrait placement. And the determination target image with the coordinates (Xb ′, Yb ′) and the coordinates (Xd ′, Yd ′) as the center are extracted at the two positions and the back side of the document.

  On the other hand, when the document is placed horizontally long, as shown in FIG. 7C, the coordinates (Xb, Yb), coordinates (Xd, Yd) and the surface of the document corresponding to the horizontally long placement are 2 and the back side of the document are extracted as determination target images centered on coordinates (Xa ′, Ya ′) and coordinates (Xc ′, Yc ′).

  After extracting the plurality of determination target images, the coordinate conversion unit 15 performs address conversion and rotation conversion on the reference image information or the plurality of determination target image information according to the reference corner position.

  In this specific example, the case where the address conversion and the rotation conversion are performed on the information of the reference image will be described. One of the information may be subjected to address conversion, and the other may be subjected to rotation conversion.

  For example, since the determination target image centered on the coordinates (Xa, Ya) on the document surface is the same as the original placement state when the original reference image is registered, the amount of address conversion and rotation conversion is zero. It becomes.

  In addition, the determination target image centered on the coordinates (Xb, Yb) on the document surface is rotated 90 degrees to the right with respect to the original placement state when the original reference image was registered. The address converter converts the X and Y coordinates so that the center coordinates (X0, Y0) of the reference image become the coordinates (Xb, Yb), and the rotation converter converts it 90 degrees to the right.

  In addition, the determination target image centered on the coordinates (Xc, Yc) on the document surface is rotated 180 degrees with respect to the original placement state when the original reference image is registered. The X and Y coordinates are converted by the address conversion unit so that the center coordinate (X0, Y0) of the X becomes the coordinate (Xc, Yc), and the rotation conversion unit performs a rotation of 180 degrees.

  The determination target image centered on the coordinates (Xd, Yd) on the document surface is rotated 90 degrees to the left with respect to the original placement state when the original reference image was registered. The address conversion unit converts the X and Y coordinates so that the center coordinates (X0, Y0) of the reference image become the coordinates (Xd, Yd), and the rotation conversion unit performs a rotation that rotates 90 degrees to the left.

  Further, since the determination target image centered on the coordinates (Xd ′, Yd ′) on the back side of the document is the same as the original placement state when the original reference image is registered, the amount of address conversion and rotation conversion is the same. Becomes zero.

  The determination target image centered on the coordinates (Xc ′, Yc ′) on the back side of the document is rotated 90 degrees to the right with respect to the original placement state when the original reference image is registered. The address converter converts the X and Y coordinates so that the center coordinates (X0, Y0) of the original reference image become the coordinates (Xc ′, Yc ′), and the rotation converter converts the image by 90 degrees to the right. Do.

  In addition, the determination target image centered on the coordinates (Xb ′, Yb ′) on the back side of the document is rotated by 180 degrees with respect to the original placed state when the original reference image is registered. The address conversion unit converts the X and Y coordinates so that the center coordinates (X0, Y0) of the reference image become the coordinates (Xb ′, Yb ′), and the rotation conversion unit performs a rotation that rotates 180 degrees.

  Further, in the determination target image centered on the coordinates (Xa ′, Ya ′), the original reference image is rotated 90 degrees to the left with respect to the original placement state when the original reference image was registered. The X and Y coordinates are converted by the address conversion unit so that the center coordinates (X0, Y0) of the image become the coordinates (Xa ′, Ya ′), and the rotation conversion unit performs a rotation that rotates 90 degrees to the left.

The paper pattern detection unit 14 collates the information of the reference image coordinate-converted in this way with all of the determination target images four by one. If any one of these is determined to be an original, it is determined that the read original is the original, and this is notified to the operation panel. On the other hand, if it is determined that the original image is not the original as a result of collating with the reference image information for all of the determination target images for each of the front and back sides, the read original is determined not to be an original, and the operation panel 17 is notified accordingly. .
In this example, the reference data of one place was collated with the determination target images at four places on the front and back sides.
When there are four reference data corresponding to the four sides of the document, each determination target image may be one. When there are two reference data corresponding to two sides of the document, the determination target images may be two in each case.

<Third Embodiment: Device Configuration>
FIG. 8 is a configuration diagram illustrating an original collating apparatus according to the third embodiment. The original collation apparatus 1 according to the third embodiment includes a reading unit 11, an AD conversion unit 12, an image processing unit 13, a paper pattern detection unit 14, a coordinate conversion unit 15, a timing generation unit 16, an operation panel 17, and an edge detection unit 25. The control unit 18 and the system controller 19 are provided.

  The reading unit 11 reads an image of a document and generates image information. For example, an image input portion of a scanner, an image input portion of a copier, or a multifunction peripheral.

  The AD conversion unit 12 is a part that performs processing for converting analog image information read by the reading unit 11 into digital.

  The image processing unit 13 is a part that inputs the image information converted into a digital signal by the AD conversion unit 12 and performs predetermined image processing. For example, the white reference is adjusted and the color tone such as gamma correction is adjusted.

  The paper pattern detection unit 14 uses information on the pattern (paper pattern) formed by the entanglement of the fibrous material of the paper from the image information, and matches the information on the read paper pattern and the original paper pattern information. This is a determination part that calculates the degree and compares the read original with the original. Specifically, as disclosed in Patent Document 1, a determination target image at a position corresponding to the reference image at a specific position of the original is extracted from the image information of the read original, and the reference image information and the determination target image are extracted. The correlation value by the normalized correlation method is calculated while shifting the above information in the X and Y directions one pixel at a time. Then, it is determined whether or not the read original is an original by comparing the correlation value or the value of the normalized score obtained from the correlation value with a predetermined threshold value.

  The coordinate conversion unit 15 is a part that performs coordinate conversion on a predetermined determination target image in the information on the original reference image registered in advance or the image information on the read document. The coordinate conversion unit 15 includes an address conversion unit 15a and a rotation conversion unit 15b. The address conversion unit 15a performs coordinate conversion of image information in the X and Y directions, and the rotation conversion unit 15b performs coordinate conversion related to rotation of the image information.

  The timing generation unit 16 is a part that provides a timing signal for synchronizing processing to the reading unit 11 and the AD conversion unit 12.

  The operation panel 17 is, for example, a touch panel display that serves as both display means and input means. The display means of the operation panel 17 also functions as a notification unit. When it is determined that the read original is an original, a notification to that effect is given. When it is determined that the original is not the original, the opposite side of the original is read. A notice to that effect and a notice to the effect that it is not the original.

  The edge detection unit 25 performs a process of detecting a side (edge) portion of the read document with respect to the image information after the image processing is performed by the image processing unit 13.

  The control unit 18 mainly gives an instruction to the image processing unit 13 and the coordinate conversion unit 15 and receives a determination result sent from the paper pattern detection unit 14 to perform a predetermined process. In particular, in the original collating apparatus 1 of the present embodiment, the control unit 18 implements an extraction unit 18a that extracts information on a plurality of determination target images set in advance from image information of a document read by the reading unit 11. .

  The extraction unit 18a of the control unit 18 extracts a determination target image at a position based on the corner of the read document. Specifically, according to the timing generated by the timing generation unit 16, an instruction to extract image information having predetermined XY coordinates is given to the image processing unit 13. As a result, the extracted image information is sent from the image processing unit 13 to the paper pattern detection unit 14.

  Further, the control unit 18 performs a process of correcting the document reading position deviation based on the position of the document edge detected by the edge detection unit 25.

  The system controller 19 is a part that controls each part, and controls the entire apparatus.

  Further, the original collating apparatus 1 according to the present embodiment includes an automatic document feeder 21 and a document size detecting unit 22 as necessary. The automatic document feeder 21 has a function of automatically feeding a document to be read to the reading unit 11. The document size detection unit 22 has a function of detecting the size of the document placed on the reading unit 11 and determining the orientation of the document.

  In the original collating apparatus 1 of the present embodiment having such a configuration, predetermined coordinate conversion is performed on the information of the reference image of the original registered in advance or the information of the plurality of determination target images extracted by the extracting unit 18a of the control unit 18. This is performed by the coordinate conversion unit 15. Further, the paper pattern detection unit 14 compares the information of the plurality of determination target images with the information of the reference image of the original registered in advance for the information subjected to the coordinate conversion by the coordinate conversion unit 15, and the read original is the original. It is determined whether or not there is.

  In making such a determination, the reading position may deviate from the actual original when there is a large misalignment when the original is placed on the reading unit 11 or when the adjustment of the reading unit 11 is deviated. Therefore, the edge detection unit 25 can detect the position of the edge of the document from the image information of the document and set the collation range with the reference image based on the detected information. Thereby, it is possible to correct the deviation of the reading unit 11 itself and the document placement deviation. Furthermore, it is possible to detect and correct document skew when the automatic document feeder 21 is used, leading to improved detection accuracy.

  Note that, as the configuration of the original collation apparatus 1 provided with the edge detection unit 25, the edge detection unit 25 is used in a configuration that can simultaneously capture image information on the front and back sides of the document as in the configuration of the second embodiment shown in FIG. You may apply. In this case, edges are detected for both the front and back image information of the document.

<Third Embodiment: Original Verification Method>
First, in accordance with an instruction on the operation panel 17, the system controller 19 reads information on the original reference image registered in the storage unit M in advance and transfers it to the control unit 18.

  Next, the reading unit 11 reads image information of a document to be collated. In addition to being placed directly at the reading position of the reading unit 11, the document may be sent to the reading position by the automatic document feeder 21.

  Next, the image information of the document read by the reading unit 11 is converted into a digital signal by the AD conversion unit 12, and color adjustment such as white reference adjustment and gamma correction is performed by the image processing unit 13. Thereafter, information of a plurality of determination target images set in advance is extracted from the image information of the document by the extraction unit 18a of the control unit 18.

  Here, the positions of the plurality of determination target images extracted by the control unit 18 are positions based on the corners of the read document. For example, when a rectangular document is read, determination target images at four positions with respect to each corner of the rectangle are extracted. Further, when a triangular document is read, determination target images at three positions with respect to each corner of the triangle are extracted. This is based on the premise that the document is placed on the reading unit 11 with a corner as a reference, regardless of which corner the document is placed on, the determination target corresponding to the position of the original reference image This is so that an image can be obtained.

  Further, when the edge detection unit 25 detects the position of the edge of the document and detects that there is a shift or inclination in the reading position of the document, the extraction unit 18a of the control unit 18 corrects the shift or inclination. The target image is extracted in the state.

  Next, the coordinate conversion unit 15 uses the reference position of the reference corner of the original reference image information registered in the storage unit M or the information of the plurality of determination target images extracted by the extraction unit 18 a of the control unit 18. Only the coordinates in the X and Y directions according to the coordinates are converted by the address converter 15a, and the coordinates are converted by the rotation converter 15b so as to rotate by an angle corresponding to the position of the reference corner.

  Next, the paper pattern detection unit 14 collates the information subjected to coordinate conversion by the coordinate conversion unit 15 with a plurality of determination target images and the original reference image. In other words, using the information on the determination target image or the reference image that has undergone coordinate transformation, all of the plurality of determination target images are checked against the original reference image, and at least one is determined to be the original If there is, it is determined that the read original is the original. If it is determined to be the original, it is notified to the operation panel 17 via the system controller 19.

  On the other hand, if all of the plurality of determination target images are not determined to be originals, it is determined that the read original is not the original, or the operation panel 17 is notified that the opposite side of the original is read. This is for avoiding this by reading the opposite side of the original again because the read side of the original may be opposite to the front and back when the original reference image is registered.

  At this time, since the user may make a mistake in reversing the document, a part of the image information at the time of collation of the previously captured document surface is stored, and the document surface temporarily stored when the image information on the back side of the document is captured It may be confirmed that the image information does not match the image information. If the automatic document feeder 21 is provided, the reading determination mode on the platen may be prohibited, and the automatic document feeder 21 may automatically reverse the document.

  Even when the opposite side of the document is scanned, multiple images to be judged are extracted in the same way as before, all of them are checked against the original reference image, and if any of them is judged to be the original, it is scanned. The original is determined to be the original. If it is determined to be the original, it is notified to the operation panel 17 via the system controller 19.

  On the other hand, if all of the plurality of determination target images are not determined to be originals, it is determined that the read original is not an original. Here, since both sides of the document have already been collated, notification that the opposite side of the document is read again is not performed, and it is determined that the document is not the original document. When it is determined that it is not the original, it is notified to the operation panel 17 via the system controller 19.

<Specific examples of deviation correction>
FIG. 9 is a schematic diagram for explaining the deviation between the position of the document and the reading position. When there is a deviation in the placement of the original or the adjustment of the reading unit, as shown in FIG. 9A, a deviation occurs between the actual original position and the reading position. In this case, since the positions of the reference image and the determination target image do not match, accurate matching cannot be performed.

  Therefore, the reading unit reads an image having a size larger than that of the original, detects the edge of the original by the edge detection unit, and corrects the position of the determination target image based on the detected information, as shown in FIG. 9B. As described above, the determination target image and the reference image can be reliably verified.

  FIG. 10 is a schematic diagram for explaining a case where a skew is generated in a document. As shown in FIG. 10A, when skew is generated in the document, image information is read in a tilted state. Therefore, when the edge detection unit detects the edge of the document, the amount of document skew can be calculated from the direction of the edge. The control unit performs angle correction according to the amount of skew and extracts a determination target image. As a result, as shown in FIG. 10B, the determination target image can be extracted in a state where the skew of the document is corrected.

  After the edge detection unit detects the edge of the document and corrects the document position deviation, reading deviation, and skew to extract the determination target image, the original is collated by the same processing as in the first embodiment.

<Original verification program>
11 to 13 are flowcharts for explaining the flow of the original collation program according to the present embodiment. The original collation program according to the present embodiment is executed by a computer, and may be executed by a computer incorporated in a device such as a scanner, a copier, or a multifunction machine applied as an original collation apparatus of the present embodiment, It may be executed by a general-purpose personal computer. The original collation program according to the present embodiment is also stored and distributed in a storage medium such as a CD-ROM or DVD-ROM, or distributed via a line such as the Internet.

  The example of the original collation program shown in FIG. 11 to FIG. 13 shows the flow of processing in the original collation apparatus according to the first embodiment shown in FIG. 1, but when executed on a general-purpose personal computer. Takes the image information of the document from the outside and executes the same processing on the image information.

  First, as shown in FIG. 11, when it is detected that a document has been set (step S11), an instruction to perform original collation processing is received from the operation panel (step S12). Next, the system controller reads the reference image recognition position (X, Y coordinates), the size of the original, and the information of the original reference image from the information of the original registered in advance. Here, the X direction in the X and Y coordinates is the main scanning direction (electrical scanning direction) in image reading by the reading unit, and the Y direction is a direction orthogonal to the main scanning direction.

  Next, the system controller calculates four types (eight types in the case of both front and back) of the X and Y coordinates according to the position of the determination target image to be extracted, sets it in the address conversion unit, and sets the original reference image Is stored in the memory of the control unit (step S13).

  Next, it is determined whether or not a document is set on the automatic document feeder (step S14). When a document is set on the automatic document feeder, the processing shown in FIG. 12 is performed. That is, document conveyance and reading of document image information are started (step S21).

  Next, the control unit performs XY counting. The reading unit photoelectrically converts the image of the document and converts it into a digital signal by the AD conversion unit, and the image processing unit executes image processing such as shading correction and mirror conversion (step S22).

  Next, the address conversion unit sends the information of the original reference image to the paper pattern detection unit at an address that matches the XY count value. At this time, since the original reference image can have four types of orientations on one side, rotation (rotation conversion) determined by each XY address value is performed (step S23).

  Next, the information of the determination target image corresponding to the XY count value of the read original is compared with the information of the original reference image after the rotation conversion (step S24). In this comparison, it is determined whether or not the original collation matches (step S25). If they match, a message indicating that the original matches is displayed (step S50).

  On the other hand, if it does not match the original, the front and back sides of the document are reversed by the automatic document feeder (step S26), and the processing from step S21 to step S24 is repeated (step S27). Then, it is determined whether or not the original collation matches (step S28). If they match, a message indicating that the original collation has been matched is displayed (step S50). On the other hand, if they do not match, a message indicating that it is different from the original is displayed (step S51).

  If it is determined in step S14 that no document is set on the automatic document feeder, the processing shown in FIG. 13 is performed. That is, reading of the image information of the document is started by scanning (step S31).

  Next, the control unit performs XY counting. The reading unit photoelectrically converts the image of the document and converts it into a digital signal by the AD conversion unit, and the image processing unit executes image processing such as shading correction and mirror conversion (step S32).

  Next, the address conversion unit sends the information of the original reference image to the paper pattern detection unit at an address that matches the XY count value. At this time, since the original reference image can have four types of orientations on one side, rotation (rotation conversion) determined by each XY address value is performed (step S33).

  Next, the information of the determination target image corresponding to the XY count value of the read original is compared with the information of the original reference image after the rotation conversion (step S34). In this comparison, it is determined whether or not the original collation matches (step S35). If they match, a message indicating that the original matches is displayed (step S60).

  On the other hand, if it does not match the original, a message is displayed to invert the document and place it on the platen (step S36). Thereafter, when a start instruction is accepted (step S37), the processes from step S31 to step S34 are repeated (step S38). Then, it is determined whether or not the original collation matches (step S39). If they match, a message indicating that the original collation is matched is displayed (step S60). On the other hand, if they do not match, a message indicating that it is different from the original is displayed (step S61).

  In each of the embodiments described above, the image information of the paper pattern of the original paper is used as the information of the original reference image. However, the present invention is not limited to this, and the original is uniquely identified from the original image. Any other information may be used.

It is a block diagram explaining the original collation apparatus which concerns on 1st Embodiment. It is a schematic diagram explaining the reference image of the original. FIG. 6 is a schematic diagram illustrating a relationship between a document placement state and a determination target image. It is a schematic diagram explaining the position of the determination target image extracted in this embodiment. It is a block diagram explaining the original collation apparatus which concerns on 2nd Embodiment. FIG. 6 is a schematic diagram illustrating a relationship between a document placement state and a determination target image on the back surface. It is a schematic diagram explaining the position of the determination target image extracted in this embodiment. It is a block diagram explaining the original collation apparatus which concerns on 3rd Embodiment. FIG. 6 is a schematic diagram illustrating a deviation between a document position and a reading position. FIG. 6 is a schematic diagram for explaining a case where skew is generated in a document. It is a flowchart (the 1) explaining the flow of the original collation program which concerns on this embodiment. It is a flowchart (the 2) explaining the flow of the original collation program which concerns on this embodiment. It is a flowchart (the 3) explaining the flow of the original collation program which concerns on this embodiment.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 1 ... Original collation apparatus, 11 ... Reading part, 12 ... AD conversion part, 13 ... Image processing part, 14 ... Paper pattern detection part, 15 ... Coordinate conversion part, 15a ... Address conversion part, 15b ... Rotation conversion part, 16 ... Timing generation unit, 17 ... operation panel, 18 ... control unit, 18a ... extraction unit, 19 ... system controller, 21 ... automatic document feeder, 22 ... document size detection unit, M ... storage means

Claims (7)

Reading means for reading image information of a document;
Extracting means for extracting information of a plurality of determination target images set in advance from image information of the original read by the reading means;
Coordinate conversion means for performing coordinate conversion on information of a reference image of an original document registered in advance or information on a plurality of determination target images extracted by the extraction means;
Judgment for comparing the information of the plurality of determination target images with the information of the reference image of the original registered in advance with respect to the information subjected to the coordinate conversion by the coordinate conversion unit, and determining whether or not the original is the original An original collating apparatus having means. The reading means reads image information on the front and back of the document,
The extraction unit extracts the plurality of determination target images from image information on the front and back sides of the document read by the reading unit,
2. The original collating apparatus according to claim 1, wherein the determination unit compares information on the plurality of determination target images on the front and back sides of the document extracted by the extraction unit with information on the reference image. Comprising edge detection means for detecting the position of the edge of the document from the image information of the document read by the reading means;
2. The original collating apparatus according to claim 1, wherein the extracting unit extracts information of the plurality of determination target images on the basis of positions of corners formed by adjacent two sides of the document detected by the side detecting unit. The original collating apparatus according to claim 1, further comprising a notifying unit configured to notify the reading unit to read image information on the opposite side of the original when the determining unit determines that the original is not an original. The original collating apparatus according to claim 1, wherein the extraction positions of the information of the plurality of determination target images are positions corresponding to the corners of the document. Reading the image information of the document;
Extracting information of a plurality of determination target images set in advance from the read image information of the document;
A step of performing coordinate conversion on information on a reference image of an original document registered in advance or information on a plurality of determination target images extracted by the extraction unit;
Comparing the information of the plurality of determination target images with the information of the reference image of the original registered in advance for the coordinate-converted information, and determining whether or not the original is the original Method. A step of capturing image information of a document;
Extracting information of a plurality of determination target images set in advance from the image information of the captured original;
A step of performing coordinate conversion on information on a reference image of an original document registered in advance or information on a plurality of determination target images extracted by the extraction unit;
A step of comparing the information of the plurality of determination target images with the information of the reference image of the original document registered in advance with respect to the coordinate-converted information, and determining whether or not the document is the original document by a computer The original collation program to be made.

Images