JP2012039494A - Image processing apparatus, image processing system and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To suppress an influence of noise in the case of deciding true or false of the medium by acquiring a pattern of a fiber formed on a surface of a medium from an image expressed by a device color space, and by comparing an image indicating the pattern with an image registered in advance.SOLUTION: A control part acquires, based on a scan image read by a scanner which reads a surface of a medium optically, an image indicating a pattern appearing in a collation area due to a fiber forming the medium that is a collation image expressed by the device color space (S204). The control part carries out smoothing of the collation image by performing smoothing processing (S205). Then, the control part compares the collation image where the smoothing is carried out with a registered image (S206). Then, the control part, by deciding whether the predetermined conditions of the original document is satisfied or not based on the comparison result, decides the true or false of the manuscript paper (S207).

Description

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

  Patent Document 1 below describes the authenticity of the medium by reading the pattern of the fiber formed on the surface of the medium (for example, paper) with a scanner and comparing the image showing the pattern with a pre-registered image. An apparatus for determining is disclosed.

Japanese Patent No. 4103826

  An object of the present invention is to acquire a pattern of a fiber formed on the surface of a medium from an image expressed in a device color space, and compare the image showing the pattern with a pre-registered image. Is to suppress the influence of noise.

  The invention according to claim 1 for solving the above-mentioned problem forms the medium on the basis of the image read by the image reading means for optically reading the surface of the medium and expressed in the device color space. A collation image acquisition unit that acquires a collation image indicating a pattern appearing in the collation region by the fibers that perform, a smoothing unit that smoothes the collation image acquired by the collation image acquisition unit by performing a smoothing process, and A comparison unit that compares the collated image smoothed by the smoothing unit with a registered image, and a determination that determines whether a predetermined original condition is satisfied based on a comparison result of the comparison unit And an image processing apparatus.

  According to a second aspect of the present invention, in the first aspect of the invention, the smoothing means includes a specifying means for specifying the type of the medium, and a spatial frequency component having a spatial frequency greater than the spatial frequency corresponding to the type of the medium. Means for smoothing the collation image acquired by the collation image acquisition means by performing a smoothing process according to the type specified by the specification means in order to reduce image quality.

  According to a third aspect of the present invention for solving the above problem, the medium is an image read by an image reading means for optically reading the surface of the medium, and is based on the image expressed in a device color space. A collation image acquiring means for acquiring a collation image showing a pattern appearing in the collation region by the fibers forming the smoothing means for smoothing the collation image obtained by the collation image acquisition means by performing a smoothing process, Comparison means for comparing the collated image smoothed by the smoothing means and the registered image, and determination means for judging whether or not a predetermined original condition is satisfied based on a comparison result of the comparison means , A program that causes a computer to function.

  According to a fourth aspect of the present invention, there is provided an image processing system including an image reading device and an image processing device, wherein the image reading device is an image showing an image of a surface of a medium. An image expressed in a device color space is read from the surface of the medium, and the image processing apparatus is configured to display a pattern that appears in a matching region by fibers forming the medium based on the image read by the image reading apparatus. A collation image acquisition means for obtaining a collation image indicating the smoothing processing, smoothing means for smoothing the collation image obtained by the collation image acquisition means by performing a smoothing process, and smoothing by the smoothing means Comparing means for comparing the collated image with the registered image; and determining means for determining whether or not a predetermined original condition is satisfied based on a comparison result of the comparing means; Include, characterized by.

  According to the first, third, and fourth aspects of the present invention, the pattern of the fiber formed on the surface of the medium is obtained from the image expressed in the device color space, and the image indicating the pattern is registered in advance. The influence of noise when determining the authenticity of the medium can be suppressed by comparing it with the image compared with the case without this configuration.

  According to the invention of claim 2, the pattern of the fiber formed on the surface of the medium is obtained from the image expressed in the device color space, and the image showing the pattern is compared with the image registered in advance. The influence of noise when determining the authenticity of a medium can be suppressed regardless of the type of medium as compared with the case without this configuration.

1 is a configuration diagram of an image processing system including an image processing apparatus according to an embodiment of the present invention. It is a figure which illustrates manuscript paper. It is a figure which illustrates the process which a control part performs. It is a figure which illustrates the data memorize | stored in a hard disk. It is a figure which illustrates the memory content of a database. It is a figure which illustrates the process which a control part performs.

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

[Image processing system]
FIG. 1 is a configuration diagram of an image processing system 1 including an image processing apparatus according to an embodiment of the present invention. The image processing system 1 includes an image processing device 2 according to an embodiment of the present invention and a database 4. The image processing apparatus 2 and the database 4 are communicably connected via a network.

[Image processing device]
The image processing apparatus 2 is, for example, a multifunction peripheral. As shown in FIG. 1, the image processing apparatus 2 is a control unit 6, a main memory 8, a hard disk 10, an operation input unit 12, a display unit 14, and a network interface 16 (hereinafter referred to as a network interface 16). , Described as “network IF”) and a scanner 18 (image reading means). In addition to these, the image processing apparatus 2 includes an image forming unit that performs printing.

  The control unit 6 is a microprocessor, for example, and executes information processing according to a program stored in the main memory 8.

  The main memory 8 is a RAM, for example. The main memory 8 stores the above program. This program is read from a computer-readable information storage medium such as a flexible disk, CD-ROM, DVD (registered trademark) -ROM, magnetic tape, hard disk, MO, MD, IC card, and stored in the main memory 8. Alternatively, it may be supplied from a communication network such as the Internet and stored in the main memory 8.

  The main memory 8 also stores various data necessary for information processing.

  The hard disk 10 is a storage medium used as an auxiliary storage device, and stores various information. The contents stored in the hard disk 10 will be described later.

  The operation input unit 12 is an interface for a user to perform an operation input, and outputs a signal indicating a user operation content to the control unit 6.

  The display unit 14 is, for example, a liquid crystal display provided on the side surface of the image processing apparatus 2 and displays information output from the control unit 6.

  The network IF 16 is an interface for connecting the image processing apparatus 2 to a network, and is a network interface card, for example. The network IF 16 outputs information received from the network to the control unit 6 and transmits information output from the control unit 6 to the network.

  The scanner 18 includes a plurality of imaging sensors (here, CCD sensors), scans the surface of the original paper (medium), and optically reads the surface of the original paper. That is, an image on the surface of the original paper is read from the surface of the original paper. The scanner 18 scans the surface of the original paper, and outputs the read image (hereinafter referred to as “scanned image”) to the control unit 6.

  Specifically, the scanner 18 performs A / D conversion on image signals sequentially output from the CCD sensors, and supplies them to the control unit 6 as pixel values of pixels included in the scan image. The A / D converted image signal itself is a pixel value of a pixel included in the scan image, and the scan image is expressed in the device color space. Here, the device color space refers to a color space depending on various devices such as a scanner and a copier. Note that the scanner 18 may perform gain adjustment corresponding to each pixel at the time of A / D conversion.

  In the case of the present embodiment, scanning is performed so that the resolution of the scanned image is 600 pdi.

[Database]
The database 4 is a database server, for example. Here, the database 4 is provided as an external storage device outside the image processing apparatus 2, but may be constructed in the hard disk 10. The significance of the database 4 will be described later.

[About scan mode]
The image processing apparatus 2 is provided with a plurality of scan modes including a print mode, a normal mode, a registration mode, and a collation mode, for example. The user selects one of these scan modes, and causes the image processing apparatus 2 to scan the original paper.

  Hereinafter, “scanning in a state where a certain scan mode X is selected” is referred to as “scanning in scan mode X”.

[Print mode, Normal mode]
The print mode is selected when the scanned image is printed. The print mode is selected when the scanned image is copied. When the scanner 18 scans the original paper in the print mode or the normal mode, when the control unit 6 acquires the scan image output from the scanner 18, the control unit 6 performs predetermined image processing on the scan image. For example, since the scan image is expressed in the device color space, image processing for converting the color space of the scan image from the device color space to the device-independent color space is executed. Here, the device-independent color space refers to a color space that does not depend on the device in which the scan image is input. Then, when the scan mode is the print mode, the control unit 6 causes the image forming unit to print the scan image based on the scan image subjected to the image processing. When the scan mode is the normal mode, the scan unit The image is stored in the hard disk 10.

[Registration mode]
The registration mode is used when registering a pattern that appears on the surface by fibers forming the original paper. For example, when selecting in the registration mode and causing the image processing apparatus 2 to perform scanning, the user causes the scanner 18 to scan the original paper illustrated in FIG. FIG. 2 is an example of original paper. A code image 20 is printed in advance on the upper left of the original paper. The code image 20 is an image obtained by encoding a paper ID that is identification information of an original paper. The meaning of the area indicated by reference numeral 22 and the area indicated by reference numeral 24 will be described later.

  Further, for example, when the user selects in the registration mode and causes the image processing apparatus 2 to perform scanning, the user also designates the material type of the original paper with the operation input unit 12. For example, the user designates a corresponding material type among predetermined material type candidates (plain paper, recycled paper, etc.).

  Under such circumstances, the scanner 18 scans the original paper in the registration mode, and the control unit 6 executes the processing illustrated in FIG. 3 according to the program stored in the main memory 8. That is, the control unit 6 acquires a scan image output from the scanner 18 (S101). As described above, the scan image is expressed in the device color space.

  Then, the control unit 6 analyzes the code image 20 included in the scanned image to obtain a paper ID that is identification information of the original paper (S102).

  And the control part 6 acquires a registration image by cutting out a registration area | region from scan images (S103). Here, the registration area is an area of a predetermined pixel range, and is an area for acquiring a registration image, and corresponds to the area 22 shown in FIG. The registered image is an image registered in association with the paper ID that is identification information of the original paper, and is used to determine the originality of the original paper.

  And the control part 6 smoothes a registration image by performing the smoothing process (S104). Thereby, the control unit 6 removes noise (for example, noise due to deterioration of each CCD sensor) caused by each CCD sensor included in the registered image.

  For example, the control unit 6 smoothes the registered image by filtering the registered image using a spatial filter. For example, the control unit 6 smoothes the registered image by performing filtering using a Gaussian filter. Further, for example, after the Fourier transform of the registered image, the control unit 6 smoothes the registered image by performing an inverse Fourier transform after deleting a spatial frequency component having a spatial frequency larger than the reference spatial frequency. For example, the control unit 6 smoothes the registered image by reducing the registered image.

  In the case of this embodiment, the control unit 6 smoothes the registered image by performing a smoothing process according to the material type specified by the user. Specifically, data illustrated in FIG. 4 is stored in the hard disk 10. As shown in FIG. 4, this data is data in which smoothing control data is associated with each material type, and the control unit 6 is based on the smoothing control data associated with the material type designated by the user. Perform smoothing.

  For example, when the registered image is smoothed by the first method, the data illustrated in FIG. 4 includes a spatial filter used for smoothing as the smoothing control data. For example, when the registered image is smoothed by the second method, the data shown in FIG. 4 includes information for determining the Gaussian filter (for example, the size of the Gaussian filter and the standard deviation value of the Gaussian function). ) Is included as smoothing control data. For example, when the registered image is smoothed by the third method, the data shown in FIG. 4 includes the reference spatial frequency as smoothing control data. For example, when the registered image is smoothed by the fourth method, the data shown in FIG. 4 includes the reduction ratio as the smoothing control data.

  Here, the smoothing control data associated with each material type is set so that a spatial frequency component having a spatial frequency higher than the reference spatial frequency corresponding to the material type is reduced. For example, when the main spatial frequency of the fiber pattern forming the original paper of a certain material X is the spatial frequency X, the smoothing control data associated with the material X is a space having a spatial frequency larger than the reference spatial frequency X. The frequency component is set to be reduced. For example, the smoothing control data associated with “plain paper” is set so that spatial frequency components having a spatial frequency greater than 200 dpi are reduced. Therefore, the feature of the fiber pattern forming the original paper is not lost by the smoothing regardless of the original paper.

  Further, scanning is performed so that the resolution of the scanned image (registered image) is higher than the reference spatial frequency of any material (that is, 600 dpi). Therefore, considering that the main spatial frequency of noise caused by the CCD sensor is around the resolution of the scan image (registered image), smoothing is performed as described above, and the scan image (registered image) By performing scanning so that the resolution is higher than the reference spatial frequency of any material, only noise caused by the CCD sensor is removed, and the feature of the fiber pattern forming the original paper is not lost. It has become.

  The control unit 6 may obtain a smoothed registered image by smoothing a scan image by performing a smoothing process and cutting out a registration area from the smoothed scan image.

  After executing step S104, the control unit 6 registers the smoothed registered image in the database 4 (S105). That is, the control unit 6 stores the smoothed registered image, the material type designated by the user, and the paper ID acquired in step S102 in the database 4 in association with each other. FIG. 5 illustrates the stored contents of the database 4.

[Verification mode]
The collation mode is used when confirming the originality (ie authenticity) of the original paper. Even when the collation mode is selected and the image processing apparatus 2 performs scanning, the user causes the scanner 18 to scan the document sheet shown in FIG. However, in this embodiment, the user does not specify the material type of the original paper.

  Under such circumstances, the scanner 18 scans the original paper in the collation mode, and the control unit 6 executes the process shown in FIG. 6 in accordance with the program stored in the main memory 8. That is, the control unit 6 acquires a scan image output from the scanner 18 in the same manner as in step S101 (S201). As described above, the scan image is expressed in the device color space.

  Then, the control unit 6 acquires a paper ID which is document paper identification information in the same manner as in step S102 (S202).

  Then, the control unit 6 (smoothing means, specifying means) specifies the material type of the original paper (S203). Specifically, the control unit 6 refers to the database 4 and acquires information indicating the material type associated with the paper ID of the original paper.

  And the control part 6 (collation image acquisition means) acquires a collation image by cutting out a collation area | region from scan images (S204). Here, the collation area is an area of a predetermined pixel range, and is an area for acquiring a collation image. This collation area is an area including a registration area, and corresponds to the area 24 shown in FIG. The collation image is an image collated with a registered image in order to determine the originality of the original paper.

  And the control part 6 (smoothing means) smoothes a collation image by performing the smoothing process similarly to the step of S104 (S205). Thereby, the control unit 6 removes noise (for example, noise due to deterioration of each CCD sensor) caused by each CCD sensor included in the collation image.

  That is, for example, the control unit 6 smoothes the collation image by performing a smoothing process according to the material type specified in step S203. Specifically, the smoothing process is executed based on the smoothing control data associated with the material type specified in step S203 of the control unit 6.

  As described above, the smoothing control data associated with each material type is set so that a spatial frequency component having a spatial frequency larger than the reference spatial frequency corresponding to the material type is reduced. Therefore, the feature of the fiber pattern forming the original paper is not lost by the smoothing regardless of the original paper.

  Further, as described above, scanning is performed so that the resolution of the scanned image (collation image) is higher than the reference spatial frequency of any material (that is, 600 dpi). Therefore, considering that the main spatial frequency of noise caused by the CCD sensor is around the resolution of the scan image (collation image), smoothing is performed as described above, and the scan image (collation image) By performing scanning so that the resolution is higher than the reference spatial frequency of any material, only noise caused by the CCD sensor is suppressed, and the feature of the fiber pattern forming the original paper is not lost. It has become.

  Note that the control unit 6 may obtain a smoothed collation image by smoothing the scan image by performing a smoothing process and cutting out a collation region from the smoothed scan image.

  After executing step S205, the control unit 6 (comparison unit) reads the registered image associated with the sheet ID from the database 4, and compares the registered image with the smoothed collation image (S206).

  Specifically, as described in Japanese Patent No. 4103826, the control unit 6 moves the registered image within the collation image, while the registered image and the portion of the collation image that overlaps the registration image. Correlation values between images (hereinafter referred to as partial images) are sequentially calculated according to the following formulas. Thus, the control unit 6 obtains a plurality of correlation values. This “plurality of correlation values” corresponds to the comparison result.

Here, F is a registered image, G is a partial image, f _i is the pixel value of each pixel of the registered image, N is the total number of pixels of the registered image (partial image), and g _i is It is the pixel value of each pixel of the partial image, f _Ave is the average value of the individual pixel values of the registered image, and g _Ave is the average value of the individual pixel values of the partial image.

  Then, the control unit 6 (determination means) determines whether or not the original condition is satisfied based on the comparison result between the registered image and the collation image in order to determine the originality (that is, authenticity) of the original paper. Determination is made (S207). Here, the “original condition” is a condition for determining that the original paper is genuine (original). When “original condition is satisfied”, it is determined that the original paper is genuine. In the present embodiment, “when the original condition is satisfied” means that “the maximum value of the plurality of correlation values obtained in step S206 is equal to or greater than a first reference value (eg, 0.3); A value obtained by dividing an average value of the plurality of correlation values from the maximum value by a standard deviation of the plurality of correlation values (hereinafter referred to as a normalized score) is a second reference value (for example, 0.5 ) If this is the case. Probability FRR (False Refection Rate) that the original paper is determined to be fake even though the original paper is genuine, and the original paper is determined to be genuine even though the original paper is fake In order to lower the probability FAR (False Acceptance Rate), the condition for satisfying the original condition is not only that the maximum correlation value is greater than or equal to the first reference value, but also that the normalized score is It is also required to be 2 or more reference values.

  In step S207, the control unit 6 determines whether or not the maximum correlation value is greater than or equal to the first reference value. The control unit 6 also determines whether or not the normalized score is equal to or less than the second reference value. By doing so, the control unit 6 determines whether or not the original condition is satisfied, that is, whether or not the original sheet is genuine.

  Note that the control unit 6 causes the display unit 14 to display information indicating whether or not the original condition is satisfied after step S207. For example, when the original condition is satisfied, the control unit 6 displays a message that the original paper is genuine on the display unit 14, and when the original condition is not satisfied, the control unit 6 displays a message that the original paper is fake. Display on the unit 14.

[Summary]
As described above, the image processing apparatus 2 compares the color space of the verification image with the registered image without converting the color space of the verification image into the device-independent color space. The processing load is reduced as compared with the case of comparing with a registered image after conversion to. In addition, when comparing with the registered image without converting the color space of the matching image, there is a concern about the influence of noise included in the matching image, but since the matching image is compared with the registered image after performing smoothing processing. Thus, the influence of noise on the determination result of the authenticity of the original paper is suppressed.

  Further, since the image processing apparatus 2 performs the smoothing process according to the material type of the original paper, the accuracy of determining the authenticity of the original paper is improved regardless of the original paper. Become.

  Further, in the image processing apparatus 2, the accuracy of determining the authenticity of the original paper is improved from the point that not only the collation image but also the registered image is expressed in the device color space.

  The embodiment of the present invention is not limited to the above embodiment.

  For example, the “medium” is not limited to paper. For example, the “medium” may be a plastic plate.

  Further, the image processing system 1 may include a plurality of image processing apparatuses 2. In this case, registration of the registered image is performed by the first image processing device 2 of the plurality of image processing devices 2, and the authenticity of the original paper is determined by the second image processing of the plurality of image processing devices 2. It may be performed by the device 2. In this case, the user may select the registration mode and cause the first image processing apparatus 2 to perform scanning, and then select the collation mode and cause the second image processing apparatus 2 to perform scanning. .

  The user may specify the material type of the original paper when selecting the collation mode and causing the image processing apparatus 2 to perform scanning. In this case, the control unit 6 (identification means) may identify the material type of the original paper based on the content designated by the user in step S203.

  DESCRIPTION OF SYMBOLS 1 Image processing system, 2 Image processing apparatus, 4 Database, 6 Control part, 8 Main memory, 10 Hard disk, 12 Operation input part, 14 Display part, 16 Network interface, 18 Scanner, 20 Code image, 22, 24 area | region.

Claims (4)

An image read by an image reading unit that optically reads the surface of a medium, and a collation image showing a pattern appearing in a collation area by fibers forming the medium is acquired based on an image expressed in a device color space Collation image acquisition means for
Smoothing means for smoothing the collation image obtained by the collation image obtaining means by performing smoothing processing;
Comparison means for comparing the collation image smoothed by the smoothing means with the registered image;
Determination means for determining whether or not a predetermined original condition is satisfied based on a comparison result of the comparison means;
An image processing apparatus comprising: The smoothing means includes
A specifying means for specifying the type of the medium;
Acquired by the collation image acquisition means by performing smoothing processing according to the type specified by the specifying means in order to reduce the spatial frequency component of the spatial frequency larger than the spatial frequency corresponding to the type of the medium Means for smoothing the collation image,
The image processing apparatus according to claim 1. An image read by an image reading unit that optically reads the surface of a medium, and a collation image showing a pattern appearing in a collation area by fibers forming the medium is acquired based on an image expressed in a device color space Collation image acquisition means for
Smoothing means for smoothing the collation image obtained by the collation image obtaining means by performing a smoothing process;
Comparison means for comparing the collated image smoothed by the smoothing means with a registered image;
Determination means for determining whether or not a predetermined original condition is satisfied based on a comparison result of the comparison means;
As a program that allows the computer to function. An image processing system including an image reading device and an image processing device,
The image reading device includes:
An image showing an image of the surface of the medium, which is expressed in the device color space, is read from the surface of the medium,
The image processing apparatus includes:
Based on the image read by the image reading device, collation image acquisition means for acquiring a collation image showing a pattern appearing in a collation region by fibers forming the medium;
Smoothing means for smoothing the collation image obtained by the collation image obtaining means by performing smoothing processing;
Comparison means for comparing the collation image smoothed by the smoothing means with the registered image;
Determination means for determining whether or not a predetermined original condition is satisfied based on a comparison result of the comparison means,
An image processing system.

Images