JP2009118394A - Documents collating apparatus, image forming apparatus, and documents collating program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To improve precision for specifying a position of a collation area for extracting a feature quantity of a random pattern from a document to be collated, thereby reducing a collation time. <P>SOLUTION: A reading section 24 reads a document image of the document and a random pattern of a sheet as read data, a reference image is read out from registered data of an original stored in a storage section 32 for each original, and image matching is performed by comparing the read out reference image with the read data. If match is made, positional information indicating the position of the collation area determined based on a reference position set in the reference image as the reference is read out from the storage section 32, and the position of the collation area 34 is specified. The feature quantity of the random pattern is extracted from the specified area, and is collated with the feature quantity of the registered data of the original. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a document collation apparatus, an image forming apparatus, and a document collation program, and more particularly to a document collation apparatus, an image forming apparatus, and a document collation program for confirming the originality of a document.

  Forged documents such as securities, various rights documents, insurance certificates, resident's certificates, birth certificates, guarantees, passports, banknotes, confidential documents, etc. and original documents such as ID cards To prevent this, use special paper and print marks and background patterns, print with special magnetic ink, mix optical semiconductors with documents, embed invisible information by special halftone printing, etc. Special treatments that cannot be counterfeited are carried out.

  On the other hand, in a highly informationized society, it is assumed that the contents included in various documents as described above are digitized and used by printing them. For this reason, a general user may obtain the digitized content via a network such as the Internet, print it with an arbitrary printer, and use the printed matter as securities or certificates.

  Applying the special processing as described above to confirm whether or not the document printed by the printer is the original is expensive and complicated. For this reason, from a predetermined area of the original paper, a feature value of a random pattern image created by the plant fiber in a state where the plant fiber forming the paper is intertwined in an indefinite shape is extracted and stored, and becomes a collation target. 2. Description of the Related Art A paper identification / collation device that confirms originality by comparing feature values of a pattern image extracted from a predetermined area of a document with stored feature values is known (for example, Patent Document 1).

As described above, when collation is performed using the feature amount extracted from the image of the predetermined area, the same place as the area from which the stored original feature quantity is extracted is used as the observation area in the document to be collated. Need to be identified. For this reason, in the paper identification collating apparatus of Patent Document 1, the corner of the paper or a mark provided in advance is used as a starting point, and a circle area centering on this starting point or a quadrangular area having this starting point as a vertex is specified as an observation area. Yes.
JP 2004-102562 A

  The present invention relates to a document collation apparatus, an image forming apparatus, and a document that can improve the accuracy of specifying the position of a collation area for extracting a random pattern feature amount from a document to be collated and reduce the collation time. The purpose is to provide a verification program.

  In order to achieve the above object, a document collation apparatus according to the present invention is an original image created using a reading unit that reads a document and outputs read data, and a sheet having a random pattern formed by entwining plant fibers. The reference image including a part of the formation area, the position information indicating the position of the collation area determined with reference to the reference position set in the reference image, and the storage means obtained from the storage unit storing the random pattern feature amount of the collation area When a reference image is compared with the read data, and the data corresponding to the reference image is included in the read data, the position of the original in the original is determined based on the position corresponding to the reference position of the original. A specifying unit that specifies a position corresponding to the collation area, and a position corresponding to the collation area from the read data based on the position specified by the specifying unit. And determining means for extracting the feature amount of the random pattern of the region, comparing the feature amount of the extracted random pattern and the feature amount of the random pattern of the original, and determining whether or not the original is the original. .

  Further, the document collating apparatus of the present invention extracts the reference image of the original stored in the storage unit and the feature amount of the random pattern from the read data obtained by reading the original by the reading unit. Can do.

  In addition, the document collating apparatus according to the present invention further includes a detecting unit that detects the image forming area of the original, and can extract the reference image of the original based on a detection result of the detecting unit.

  Further, the document collating apparatus according to the present invention can be provided in the image forming apparatus.

  In the document collation program of the present invention, the computer is set to the reference image including a part of the original image forming region created using a sheet having a random pattern formed by intertwining plant fibers, and the reference image. Position information indicating the position of the collation area determined with reference to the reference position, the reference image obtained from the storage means storing the random pattern feature amount of the collation area, and read data output from the reading means for reading the document When the data corresponding to the reference image is included in the read data, the position corresponding to the collation area in the document is determined based on the position corresponding to the reference position of the document. Based on the specifying means for specifying and the position specified by the specifying means, a lander of an area corresponding to the collation area from the read data Extracts the feature quantity of the pattern, collates the feature quantity of the extracted random pattern with the feature quantity of the random pattern of the original, and functions as a judging means for judging whether or not the original is the original. is there.

  As described above, according to the document collating apparatus according to the first aspect, the accuracy of specifying the position of the collation area for extracting the feature quantity of the random pattern from the original to be collated is improved, and the collation time is shortened. The effect that it can be obtained.

  According to the document collating apparatus of the second aspect, it is possible to obtain the effect that the original random pattern feature quantity can be stored and the original can be read by one apparatus.

  According to the document collation apparatus according to claim 3, it is possible to appropriately extract a reference image serving as a reference for identifying a collation area, and to improve the accuracy of identifying the position of the collation area. An effect is obtained.

  According to the image forming apparatus of the fourth aspect, the accuracy of specifying the position of the collation region for extracting the feature quantity of the random pattern from the original to be collated is improved, and the collation time is shortened to form an image. The effect that it can be obtained.

  According to the document collation program according to claim 5, it is possible to improve the accuracy of specifying the position of the collation area for extracting the feature quantity of the random pattern from the manuscript to be collated, and to shorten the collation time. An effect is obtained.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. In the present embodiment, a case where the document collation apparatus 10 of the present invention is provided in the multifunction machine 12 will be described.

  FIG. 1 is a block diagram showing the configuration of the present embodiment. The multifunction machine 12 includes a document collating apparatus 10, a scanner 14, an output unit 16, and a multifunction machine control unit 18.

  The scanner 14 is connected to the multifunction machine control unit 18 via an interface. The scanner 14 reads a document image set on a document table, creates image data including characters and images for the read image, and transmits the image data to the multifunction device control unit 18. Here, an automatic document feeder (ADF) can also be applied as the document table.

  The output unit 16 feeds recording paper, an exposure apparatus composed of optical components such as a laser diode, a polygon mirror, and a lens (not shown), an electrophotographic process device such as a photoreceptor, a charger, and a developing device. A multi-function peripheral control unit having an interface, a feeding unit for feeding, a transporting device for transporting the fed recording paper, and a recording paper transporting unit including a discharge unit for discharging the recording paper on which an image is formed. 18 is connected.

  The output unit 16 oscillates a laser beam from the laser diode of the exposure apparatus based on the image data transmitted from the multi-function device control unit 18 and irradiates the photoconductor uniformly charged by the charger. Forming an electrostatic latent image, developing toner by supplying toner from a developing device, and then transferring a toner image including predetermined characters or images to a recording paper supplied from a paper feeding unit; Record.

  The multifunction device control unit 18 receives the image data transmitted from the scanner 14 and transmits the received image data to the output unit 16. The multifunction device control unit 18 controls input / output of various information to / from each unit constituting the multifunction device 12, driving of the multifunction device 12, and the like.

  The document collation apparatus 10 includes a display operation unit 20, a control unit 22, a reading unit 24, and a storage unit 32.

  The display operation unit 20 uses a display device with a touch panel, and displays a result of an original collation process described later. In addition, the display operation unit 20 displays buttons as objects made up of characters and graphic images indicating the functions of the multifunction machine 12 and the like. These buttons function as input means, and input for performing various operations such as setting of functions can be performed by selecting the buttons with a user's finger or the like. The display operation unit 20 is connected to the control unit 22.

  In the present embodiment, a display device with a touch panel is used as the display operation unit 20, but an input using a pointer that is moved by operating a mouse on the screen may be used. Also, hard keys such as a numeric keypad and a start key may be provided, and input may be performed by pressing the hard key.

  The control unit 22 includes a ROM (Read Only Memory) as a storage medium in which various programs such as an original registration processing routine and a collation processing routine described later, parameters, and the like are stored, a CPU (Central Processing Unit) that executes various programs, And a RAM (Random Access Memory) used as a work area when the CPU executes various programs.

  FIG. 2 is a schematic diagram illustrating the configuration of the reading unit 24. The reading unit 24 is provided in the sheet conveyance path of the multifunction machine 12, and an illumination unit 26 that irradiates light to the sheet, and a light receiving unit 28 that receives the transmitted light that is irradiated by the illumination unit 26 and transmitted through the sheet. It is comprised including. In addition, a window 30 is provided in the conveyance path corresponding to the position where the reading unit 24 is provided. The reading unit 24 irradiates light from the illumination unit 26, and receives light transmitted through the window 30 and the paper by the light receiving unit 28, thereby reading a document image and a random pattern of the paper.

  As shown in FIG. 2, the illumination unit 26 includes a light source 26B that outputs light, and an optical waveguide optical system 26A that guides light emitted from the light source 26B in the paper direction. For example, an LED, a halogen lamp, a fluorescent lamp, or a xenon discharge tube can be used as the light source 26B. Further, instead of the optical waveguide optical system 26A, a condensing lens that condenses light in the paper direction can also be used.

  The light receiving unit 28 is an image sensor 28B that is an area sensor composed of a CCD or the like in which a number of light receiving elements that output a signal corresponding to the amount of received light are arranged, and transmitted light that has been irradiated by the illumination unit 26 and transmitted through the paper. The lens unit 28A forms an image on the light receiving surface of the image sensor 28B, and a signal processing circuit (not shown) that converts the signal output from the image sensor 28B into digital data and outputs the digital data.

  The paper is made of a vegetable fiber material, and the entanglement of the plant fibers is random as shown in FIG. The reading unit 24 reads the shading information indicated by the random pattern formed by entwining the plant fibers as the feature information indicating the unique characteristics of the paper.

  The storage unit 32 stores a reference image read by the reading unit 24 and used for image matching, which will be described later, position information, and feature amounts of a random pattern of the paper. For example, a hard disk drive having a large-capacity storage area, etc. The non-volatile memory. The original is registered by the storage in the storage unit 32. The storage unit 32 may be provided directly in the document collation apparatus 10, or may be provided in an original management server as an external apparatus and connected to the document collation apparatus 10 via a network.

  Next, with reference to FIG. 4 and FIG. 5, the original registration processing routine of the present embodiment will be described. This process is started when a menu for original registration is selected by the operation of the display operation unit 20 by the user.

  In step 100, it is determined whether or not the original to be registered is set at a predetermined set position. This determination may be made by detecting that the original is set at a predetermined set position by a sensor, or whether an input signal input by the user pressing the start key of the display operation unit 20 is input. You may make it judge by no. If the original is set, the process proceeds to step 102. If not, the process waits until it is set.

  In step 102, the original is conveyed from the set position to the reading unit 24, and the original document image and the random pattern of the paper are read by the reading unit 24. The original document image read by the reading unit 24 and the random pattern of the paper are temporarily stored in the RAM as a work area as read data.

  Next, in step 104, it is determined whether or not the position of the region (collation region) 34 for extracting the feature amount of the random pattern from the read data is determined. In the area (image forming area) 36 where image formation is performed in the paper, the random pattern may not be read properly due to the influence of ink, toner, or the like. Therefore, as shown in FIG. The position is preferably an area where image formation is not performed, that is, a blank area of the paper. The collation area 34 is determined by, for example, displaying the read document image on the display operation unit 20 and inputting a desired area from the display operation unit 20 by a touch operation, or by determining the distance from the upper left corner of the paper, for example. It can be determined by inputting with the numeric keypad of the display operation unit 20. If an input signal by this input is input, the determination is affirmative and the routine proceeds to step 106. If not, wait until it is entered.

  In step 106, the feature quantity of the random pattern in the collation area 34 determined in step 104 is extracted from the read data stored in the RAM, and temporarily stored in the RAM.

  Next, in step 108, processing for extracting a reference image to be used for image matching at the time of collation processing described later is executed. FIG. 6 is a flowchart showing a processing routine of reference image extraction processing.

  In step 200, a window 42 of a predetermined shape (for example, a quadrangle, a sector, a triangle, etc.) having a right angle portion is provided, and the vertex of the right angle portion of the window 42 is aligned with a predetermined corner (for example, the upper left corner) of the paper. The window 42 is set for the read data so that the two sides forming the right-angled part coincide with the two sides of the sheet. Starting from the vertex of the right-angled portion of the window 42, the window 42 is enlarged at a predetermined pixel pitch while maintaining the similar shape. As a result, the window 42 is enlarged in the direction of the arrow in FIG.

  Next, in step 202, edge extraction is performed based on the read image data in the window 42. If an edge is extracted, it is determined that a part of the image forming area is included in the window 42, and step 204 is performed. If no edge is extracted, the process returns to step 200 and the window 42 enlargement process is continued until the edge is extracted.

  In step 204, the window 42 is further enlarged while maintaining the similar shape. When the window 42 is enlarged until the extracted edge coincides with the center of the window 42, the expansion of the window 42 is stopped.

  Next, in step 206, the image in the window 42 when the enlargement of the window 42 is stopped is cut out as a reference image, temporarily stored in the RAM, and the process returns.

  Next, the process proceeds to step 110 in FIG. 4, and position information used when specifying the position of the matching area 34 is calculated in the matching process described later. The position information is set, for example, by setting the diagonal point of the starting point 38 of the window 42 when the reference image is cut out as the reference position 46 of the reference image, and the window 42 when the reference image is cut out with the reference position 46 as the origin. 1 side is defined as the x-axis direction, and one orthogonal side is defined as the y-axis direction, and the coordinates (x, y) are expressed by the x-axis direction from the reference position 46 to the base point 44 of the verification region 34 and the distance in the y-axis direction. Can be calculated.

  Next, in step 112, the random pattern feature quantity, reference image, and calculated position information stored in the RAM are stored in the nonvolatile storage unit 32 as original registration data, and the process is terminated. . The storage unit 32 stores original registration data for each original in the format shown in FIG. 7, for example.

  Next, with reference to FIG. 8, an original collation processing routine will be described. This process is started when an original collation menu is selected by the operation of the display operation unit 20 by the user.

  In step 300, it is determined whether or not the document to be verified has been set at a predetermined setting position. This determination may be made by detecting with a sensor that the document has been set at a predetermined setting position, or whether an input signal input by the user pressing the start key of the display operation unit 20 is input. You may make it judge by no. If the document is set, the process proceeds to step 302. If the document is not set, the process waits until it is set.

  In step 302, the original is conveyed from the set position to the reading unit 24, and the original image of the original and the random pattern of the paper are read by the reading unit 24. The original image of the original read by the reading unit 24 and the random pattern of the paper are temporarily stored in the RAM as a work area as read data.

  Next, in step 304, one reference image is read out from the original registration data stored for each original in the storage unit 32. For example, when the registration data illustrated in FIG. 7 is stored in the storage unit 32, the reference image 1 is read as the first reference image.

  Next, in step 306, a window 42 having the same size as the window 42 when the reference image 1 is cut out is set in a portion corresponding to any one corner of the sheet of read data. Image matching is performed by matching the points of the reference image corresponding to the starting point 38 of the window 42 when the image is cut out and comparing the read data in the window 42 with the reference image 1.

  Next, in step 308, it is determined whether or not the comparison in step 306 matches. If they do not coincide with each other, the process proceeds to step 310, and it is determined whether the image matching between the data at all four corners of the sheet on the read data and the reference image 1 has been completed. If image matching with other corner data has not been performed yet, the process proceeds to step 312. Note that whether or not they match is determined not only when the data completely match, but also when the similarity indicating the degree of similarity of the data is equal to or greater than a predetermined value. good.

  In step 312, the comparison point between the read data in the window 42 and the reference image is changed by setting the window set by matching the starting point 38 of the window 42 to any one of the corners, and the change is made. Rotate the reference image according to the location. For example, when the comparison location is changed from the upper left corner to the upper right corner, the reference image 1 is rotated 90 ° to the right, and the process returns to step 306 to set the window 42 at the upper right corner, The image matching is performed by comparing the read data and the rotated reference image 1. If they do not match in step 308, the process is repeated until all four places are completed. If all four places are finished, the determination in step 310 is affirmative and the process proceeds to step 314.

  In step 314, it is determined whether image matching between the reference image and the read data has been performed for all of the original registration data registered in the storage unit 32. In the case of the registered data shown in FIG. 7, since there is still data that has not been subjected to image matching, the result is negative and the process returns to step 304 to read the reference image 2 as the next reference image. Perform image matching. In this way, the reference image is sequentially read out and image matching with the read data is performed. If the reference image matches the corresponding portion of the read data, the result is affirmative in step 308 and proceeds to step 316 to perform image matching up to the reference image n. If they do not match, the result is affirmative in step 314 and the process proceeds to step 326.

  In step 316, the position information stored in the storage unit 32 corresponding to the reference image corresponding to the corresponding portion of the read data is read from the storage unit 32 and the collation area 34 is specified. For example, when the reference image 2 coincides with the corresponding portion of the read data, (x2, y2) is read as the position information, and from the position of the reference position 46 of the reference image, x2 in the x-axis direction and y2 in the y-axis direction. Is identified as the base point 44 of the collation area 34, and a rectangular area within a predetermined range having the base point 44 as a vertex is identified as the collation area 34.

  Next, in step 318, the feature amount of the random pattern of the collation area 34 specified in step 316 is extracted from the read data.

  Next, in step 320, the feature amount stored in the storage unit 32 corresponding to the reference image that matches the corresponding portion of the read data, for example, the feature amount 2 in the above example, is read from the storage unit 32 and read. The feature quantity is collated with the feature quantity extracted in step 318.

  Next, in step 322, it is determined whether or not the collation in step 320 matches. If they match, the process proceeds to step 326, and if they do not match, the process proceeds to step 324. The case where they do not coincide with each other is a case where the reference images are the same although they are separate originals. Note that whether or not they match is determined not only when the feature amounts completely match, but also when the similarity indicating the degree of similarity of the feature amounts is equal to or greater than a predetermined value. May be.

  In step 324, it is determined whether or not all of the original registration data stored in the storage unit 32 have been collated. If there is original registration data that has not been collated yet, the process returns to step 304 and repeats from image matching between the reference image and the read data. If collation has been completed for all registered data, the process proceeds to step 326.

  In step 326, the collation result is displayed on the display operation unit 20. If the result in step 322 is affirmative and the process proceeds to step 326, a message indicating that the collated original is not different from the original is displayed. If the result in step 314 and step 324 is affirmative and the process proceeds to step 326, the collated original Is not the original, or a message indicating that the original is not registered is displayed, and the process ends.

  In this way, image matching of a reference image including a part of the image forming area is performed, and the position of the matching area is specified with reference to the reference position set in the reference image. The position of the collation area can be appropriately specified regardless of the direction in which it is performed. Further, compared with the case of specifying the corner of the paper or a mark provided in advance as a reference, the position of the collation area can be appropriately specified because it is less affected by the paper itself or the mark portion being damaged or dirty. Therefore, the accuracy of specifying the position of the collation area can be improved and the collation time can be shortened.

  In addition, by providing a detection means for detecting the image forming area in order to extract the reference image including a part of the image forming area at the time of registering the original, a more appropriate reference image can be extracted. Therefore, the accuracy of specifying the position of the collation region can be further improved.

  In the present embodiment, the reference image is extracted from the read data obtained by reading the original with the reading unit 24 at the time of registration. However, the present invention is not limited to this. For example, print data including image data, text data, command data, and the like transmitted from a personal computer connected via a network is developed as raster data that is a drawing image, and read data is used as a reference. An image may be extracted. Accordingly, when the image forming is performed by the output unit 16 of the multifunction machine 12 based on the transmitted print data, the original document can be registered by the document collating apparatus 10.

  Further, in this embodiment, when the window 42 is enlarged to cut out the reference image, the window 42 is enlarged while maintaining the similar shape until the extracted edge coincides with the center of the window 42. The present invention is not limited to this as long as an appropriate range can be cut out as a reference image used for image matching in the matching process.

  In this embodiment, an example in which the window 42 is enlarged with the corner of the paper as the starting point 38 has been described. However, as shown in FIG. 9A, the starting point is a central point on one predetermined side of the paper. As shown, the window may be enlarged in the direction of the arrow in the figure while maintaining a similar shape. In this case, in the image matching, first, a window 42 is set at a portion corresponding to the center of one predetermined side of the paper (for example, the center of the left side as shown in FIG. 9A), and reading in the window is performed. This is done by comparing the data with a reference image. If they do not match, the reference image is rotated by 90 ° in a predetermined direction, for example, to the right, and a window 42 is set at a portion corresponding to the center of the upper side of the paper to perform image matching. As described above, the reference image is sequentially rotated with respect to adjacent sides such as the next right side of the paper and the next lower side of the paper, and image matching is performed at a total of four locations. Further, as shown in FIG. 5B, the window may be enlarged with the similar shape maintained in the direction of the arrow in the figure starting from the center of the sheet. In this case, the image matching is performed by setting a window 42 at a corresponding portion in the center of the sheet and comparing the read data in the window with the reference image. If they do not match, the reference image is rotated 90 ° to the right, for example, and a window 42 is set in the corresponding portion at the center of the paper to perform image matching. In this way, image matching is performed in all four directions at the corresponding portion at the center of the sheet.

  In the present embodiment, the collation area 34 is a quadrangular area having the base point 44 as a vertex, but the present invention is not limited to this. For example, it may be a circular area centered on the starting point 44.

  In this embodiment, the example in which the position information is calculated as the xy-axis coordinates with the reference position set in the reference image as the origin has been described. However, the present invention is not limited to this, and the reference position set in the reference image is not limited to this. Any reference can be used as long as the position of the collation region can be specified as a reference.

  In the present embodiment, an example in which reading is performed using transmitted light has been described as the reading unit 24 of the image collating apparatus 10. However, reading may be performed using reflected light.

It is a block diagram which shows the structure of this Embodiment. It is a schematic block diagram which shows an example of the reading part of this Embodiment. It is an image which shows an example of the random pattern of the fiber of paper. It is a flowchart which shows the processing routine of the original registration of this Embodiment. It is the schematic which shows an example of reading data. It is a flowchart which shows the process routine of a reference | standard image extraction process. It is a figure which shows an example of the registration data of an original. It is a flowchart which shows the processing routine of collation of this Embodiment. It is a diagram for explaining another example of window setting.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Document collation apparatus 12 Multi-function machine 14 Scanner 16 Output part 18 Multi-function machine control part 20 Display operation part 22 Control part 24 Reading part 26 Illumination part 26A Optical waveguide optical system 26B Light source 28 Light-receiving part 28A Lens unit 28B Image sensor 30 Window part 32 Storage unit 34 Collation area 36 Image formation area 38 Window origin 42 Window 44 Collation area base point 46 Reference position

Claims (5)

Reading means for reading a document and outputting read data;
A reference image including a part of an original image forming area created using a paper having a random pattern formed by entangled plant fibers, and a position of a matching area defined with reference to a reference position set in the reference image The reference data obtained from the storage means storing the positional information and the feature quantity of the random pattern of the collation area are compared with the read data, and the read data includes data corresponding to the reference image. Specifying means for specifying a position corresponding to the collation area in the document with reference to a position corresponding to the reference position of the document.
Based on the position specified by the specifying means, the feature amount of the random pattern in the region corresponding to the collation region is extracted from the read data, and the extracted random pattern feature amount and the original random pattern feature amount, Judging means for judging whether or not the original is the original,
Document verification device including   The document collating apparatus according to claim 1, wherein the reference image of the original stored in the storage unit and the feature amount of the random pattern are extracted from read data obtained by reading the original by the reading unit.   The document collating apparatus according to claim 2, further comprising a detecting unit that detects the image forming area of the original, and extracting the reference image of the original based on a detection result of the detecting unit.   An image forming apparatus comprising the document collating apparatus according to claim 1. Computer
A reference image including a part of an original image forming area created using a paper having a random pattern formed by entangled plant fibers, and a position of a matching area defined with reference to a reference position set in the reference image The reference image obtained from the storage means storing the positional information and the feature quantity of the random pattern of the collation area is compared with the read data output from the reading means for reading the document, and the reference data is included in the read data. Specifying means for specifying a position corresponding to the collation area in the document with reference to a position corresponding to the reference position of the document when data corresponding to the image is included;
Based on the position specified by the specifying means, the feature amount of the random pattern in the region corresponding to the collation region is extracted from the read data, and the extracted random pattern feature amount and the original random pattern feature amount, Judging means for judging whether or not the original is the original,
Document collation program to make it work.