JP2006293464A - Security discrimination apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a security discrimination apparatus that reads a part of a serial number indicative of an issuer printed on a security with an OCR device and discriminates the security according to a discrimination level set for the issuer indicated by the serial number. <P>SOLUTION: Forgery of securities is detected by hues slightly different depending on issuers which issue securities. A discrimination level for forgery detection is therefore selected according to the first alphabet letter Pb of a serial number Pa read by the OCR device 3. For example, for securities issued in Italy, which have the first alphabet letter Pb, "S" of the serial number Pa, a discrimination table TS10 enclosed by a broken line is used. If the type and feed state of the security fed is, for example, 10 euros/face/normal, a discrimination table 6 is selected. Similarly, as for 10 euros/face/inverted, 10 euros/back/normal and 10 euros/back/inverted, discrimination tables 22, 38 and 54 are selected respectively. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a securities discriminating apparatus that discriminates, for example, securities printed in different countries or regions that are of the same type (hereinafter referred to as ticket types).

As a securities discriminating device, from the data generated by extracting the feature amount from the detection output of the bill to be discriminated securities detected by a color camera or a magnetic sensor, and the genuine bill (genuine bill) of the bill Similarly, banknotes are discriminated by comparing them with discrimination level data generated by extracting feature amounts (see, for example, Patent Document 1). The discrimination level data at this time is provided with a threshold in a predetermined range consisting of an upper limit and a lower limit in consideration of bill manufacturing errors and quality changes accompanying circulation.
Japanese Patent Laid-Open No. 5-78074 (first page, FIG. 1)

  However, in Europe, for example, banknotes for each country were printed and managed in their own country before the euro integration, but after the euro integration, banknotes of the same specification were manufactured at the printing shops in each country. But it may not show exactly the same features. For example, there is a case in which a subtle difference in the color of banknotes or a true / false feature that cannot be visually confirmed is distributed in a state of having a different genuine note distribution at each printing station. When banknotes are determined by the banknote determination device in such a distribution environment, it is necessary to set a determination level wider than the genuine note distribution generated for each printing station in order to determine the same determination level. If the discrimination level is expanded in this way, there is a problem that the possibility of discriminating a counterfeit ticket (hereinafter referred to as a counterfeit note) as a genuine note increases. In addition, there is a problem that the ability to discriminate between the correct ticket (recyclable ticket) and the damaged ticket (non-reflowable ticket) is deteriorated.

  The present invention has been made to solve the above problems, and a part of a symbol (hereinafter referred to as an issue number) that identifies an issuing place printed on a banknote is read by an OCR device. An object of the present invention is to provide a securities discriminating apparatus with good discrimination performance by discriminating the banknote using the discrimination level set for each issuing place indicated by the read serial number.

  In order to achieve the above object, the securities discriminating apparatus according to claim 1 of the present invention pre-determines the authenticity discrimination level for discriminating the authenticity of the securities for each type, each transport state, and each issuing place. Stored authenticity determination level storage means, OCR means for reading a symbol indicating the issue place printed on the securities, type detection means for detecting the type and transport state of the securities, Based on authenticity information acquisition means for acquiring authenticity information for performing authenticity determination, an issuing office corresponding to the symbol read by the OCR means, and the type and transport state detected by the type detection means. Selection means for selecting a true / false determination level from a false determination level storage means, and authenticity information acquired by the authenticity information acquisition means is compared with a true / false determination level selected by the selection means. It characterized characterized by having a a authenticity discrimination means for discriminating the authenticity of the security by the.

  In addition, the securities discriminating apparatus according to claim 3 of the present invention stores in advance the fitness judgment level for judging the security of the securities for each type, each transport state, and each issue place. A discrimination level storage unit; an OCR unit that reads a symbol indicating an issuing place printed on the securities; a type detection unit that detects a type and a conveyance state of the securities; From the damage determination level storage means based on the damage information acquisition means for acquiring the damage information, the issuing place corresponding to the symbol read by the OCR means, and the type and transport state detected by the type detection means A selection means for selecting a damage determination level, and the value obtained by comparing the damage information acquired by the damage information acquisition means with the damage determination level selected by the selection means. It characterized by having a a fitness determination means for determining a positive loss of ticket.

  According to the present invention, a part of the symbol indicating the issuing place printed on the securities is read by the OCR device, and the securities are discriminated using the discrimination level set for each issuing office indicated by the read serial number. By doing so, it is possible to provide a securities discriminating apparatus with good discrimination performance.

  Embodiments of the present invention will be described below with reference to the drawings.

  FIG. 1 is a schematic front view of a banknote discriminating apparatus 100 as an example of a securities discriminating apparatus according to Embodiment 1 of the present invention. FIG. 2 is a block diagram illustrating a configuration of the banknote discriminating apparatus 100. Hereinafter, the configuration of the bill discriminating apparatus 100 will be described with reference to FIGS. 1 and 2. In the first embodiment, a case where euro banknotes (hereinafter referred to as banknotes) P are discriminated by the banknote discriminating apparatus 100 will be described.

  The banknote discriminating apparatus 100 is provided with a supply unit 2 for supplying banknotes P in a lump. The banknotes P supplied to the supply unit 2 are taken out one by one at regular intervals by take-out rollers (not shown) and supplied to the transport path 1. The conveyance path 1 is composed of a conveyance roller (not shown) and a conveyance belt (not shown), and conveys the banknote P in the direction indicated by the arrow A from the conveyance direction upstream to the downstream. A bill detector 110 including a plurality of bill detectors is arranged along the transport path 1 from the supply unit 2 toward the downstream in the transport direction. Further, a detection processing device 7 is provided which comprehensively detects the detection result from the banknote detection unit 110 and discriminates a genuine note / fake note. Furthermore, a sorting / stacking device (stacking means) 8 for sorting and stacking the banknotes P to be conveyed based on the determination result of the detection processing device 7 is provided.

  The banknote detection unit 110 detects physical characteristics and printing characteristics of the banknote P, and includes a timing sensor 10, an OCR device (OCR means) 3, a shape detection device (shape detection means) 4, a ticket type detection device (ticket). Species detection means) 5 and authenticity detection device (authentication information acquisition means) 6.

  The timing sensor 10 is composed of an optical sensor having a projector 10a and a light receiver 10b, and detects the timing at which the optical axis between the projector 10a and the light receiver 10b is blocked by the conveyed bill P. Based on the detection result, detection of the banknote P by each of the detection devices is started.

  The OCR device 3 illuminates the upper surface of the bill P to be conveyed and illuminates the upper monochrome reflection optical system 3a for detecting the reflected light by a monochrome CCD sensor (not shown), the lower surface of the bill P being conveyed, and its reflection. It comprises a lower monochrome reflection optical system 3b for detecting light by a monochrome CCD sensor and a character recognition device 3c. The OCR device 3 is a device that reads a serial number printed on the front or back surface of the banknote P to be conveyed by the monochrome reflection optical system 3a or the monochrome reflection optical system 3b. At this time, the reflected light in the area of a predetermined serial number (hereinafter referred to as a character reading area) set by the type of the bill P (hereinafter referred to as a bill type) is read, and characters are cut out from the read image. Characters are recognized by the character recognition device 3c.

  The shape detection device 4 is a device that includes a transmission optical system 4a and a shape detection processing unit 4b, and detects the shape of the bill P being conveyed. The transmission optical system 4a detects the transmission image of the conveyed banknote P, and outputs the detection result to the shape detection processing unit 4b. The shape detection processing unit 4b measures the vertical length (vertical size) and the horizontal length (horizontal size) of the banknote P to be conveyed from the input transmission image, and sends the measurement result to the detection processing device 7. Send.

  The ticket type detection device 5 illuminates the upper surface of the bill P to be transported and illuminates the upper reflection optical system 5a for detecting the reflected light by the CCD sensor, the lower surface of the bill P to be transported, and the reflected light to the CCD sensor. It comprises a lower reflection optical 5b that is detected in step 1 and a ticket type detection processing unit 5c. The upper reflection optical system 5a and the lower reflection optical system 5b read an image in the ticket type detection area set for detecting the ticket type of the banknote P, and output it to the ticket type detection processing unit 5c. The ticket type detection processing unit 5c generates a pattern generated from the images output from the upper and lower reflection optical systems 5a and 5b from a genuine note and stores the standard pattern for each ticket type stored in a memory (not shown). Compare with (dictionary). As a result of this comparison determination, the type of bill P to be transported and the transport state (front / back, upright / inverted) are detected. By this detection, the conveyance state in the four directions of front / upright (front), front / invert (reverse), back / upright (back) and back / invert (reverse) is detected.

  In addition, in order to determine the bill type of the banknote P, the shape of the banknote P and the pattern of the banknote P need to match the true bill. The shape of the paper P is detected by the shape detection device 4. The pattern of the banknote P is the design of the banknote P, and it is not necessary to detect the color of the banknote P in order to detect this design. Therefore, as the illumination light source (not shown) for illuminating the banknote P, a monochromatic light source such as IR (infrared) or R (red) that best represents the pattern of the banknote P to be detected is used.

  The authenticity detection device (authentication information acquisition means) 6 illuminates the upper surface of the conveyed bill P, and IR (infrared), R (red), G (green), and B (blue) in the reflected light. ) The upper color reflection optical system 6a that detects the component with a color sensor, illuminates the lower surface of the bill P to be conveyed, and IR (infrared), R (red), G (green), B (in the reflected light) It comprises a lower color reflection optical system 6b for detecting a blue component by a color sensor and a true / false detection processing unit 6c. The upper color reflection optical system 6a and the lower color reflection optical system 6b read the image in the authenticity detection area set for detecting the authenticity of the bill P, and output it to the authenticity detection processing unit 6c.

  This authenticity detection area may be set not only for each ticket type but also for each transport state because the ticket size (vertical size, horizontal size) may differ depending on the ticket type. For example, the reflective optical system to be detected is reversed between when the banknote P is conveyed in the front and when it is conveyed in the reverse. Similarly, when the banknotes P are conveyed in the front and the reverse, or when the banknotes are conveyed in the reverse and the reverse, the detection area is targeted with respect to the conveyance direction. If it is not set, the detection position from the front end in the transport direction is different. From the above, the authenticity detection area is set for each ticket type and each transport direction of each ticket type.

  The authenticity detection processing unit 6c generates a feature amount by calculating the total density value of the reflected images output from the upper color reflection optical system 6a and the lower color reflection optical system 6b, and uses the feature amount for each ticket type. Comparison is made with the feature value stored for each and every transport state. As a result of this comparison / determination, the authenticity of the banknote P to be conveyed is detected.

  The detection processing device 7 includes, for example, a built-in CPU (Central Processing Unit) 7a, a memory (RAM) 7b, a memory (ROM) 7c, and an interface 7d with each of the detection devices. Control and various processes. That is, the timing sensor 10, the OCR device 3, the shape detection device 4, the ticket type detection device 5, and the authenticity detection device 6 are executed by executing a control program or a processing program stored in advance in the memory (ROM) 7c. The authenticity determination of the banknote P conveyed based on the detection result by is performed. As a result of this authenticity determination, the banknote P determined to be a genuine note is stacked on a plurality of stacking units 8c to 8h provided in the sorting and stacking device 8 for each ticket type, or for each ticket type and transport state. Moreover, the banknote P determined to be a fake ticket is accumulated in the rejected ticket accumulating unit 8 b provided in the sorting and accumulating device 8.

  FIG. 3 is a diagram for explaining reading of the serial number Pa of the banknote P by the OCR device 3. The character reading area Eo for designating the character reading range of the banknote P has a different shape depending on the ticket type, so the range is set to horizontal (X) 256 pixels and vertical (Y) 256 pixels regardless of the ticket type. Has been. The image in the character reading area Eo set in this way is read and stored in a memory (not shown). The resolution per pixel is set to a value suitable for character recognition according to the optical magnification of the OCR device. Therefore, the addresses at the four corners of the reading area Eo using the X and Y are as follows.

The upper left address is: (0, 0)
The upper right address is: (255, 0)
The lower left address is: (0, 255)
The lower right address is: (255, 255)
The serial number Pa is composed of 12-digit characters, and the first English letter Pb of the serial number Pa represents the issuing place.

  The serial number Pa configured as described above is detected and cut out by the character recognition device 3c. The first English character Pb cut out by the detection cut-out of the serial number Pa is compared with the dictionary 3d stored in the memory inside the character recognition device 3c and recognized. In the illustrated example, the first English letter Pb is “S” and is compared in order with the letters “Z”, “Y”, “X”, “V”,... Stored in the dictionary 3d. Being recognized. In the illustrated example, the first English letter Pb is “S”, indicating that the banknote is issued in Italy.

  FIG. 4 is a diagram illustrating the shape detection of the banknote P by the shape detection device 4. The shape detection area Es that designates the shape detection range of the banknote P is set to horizontal (X) 256 pixels and vertical (Y) 256 pixels. When the bill P is conveyed in the direction indicated by the arrow (see FIG. 1) A, a transmission image in the set shape detection area Es is read. In this transmitted image, a portion that is not shielded from light is bright, and a portion that is shielded from light is dark. The resolution per pixel is set to a value suitable for shape detection according to the optical magnification of the shape detection device 4.

In order to measure the length of the banknote using the transmitted image read in this way, areas A, B, C, and D corresponding to the four corners of the banknote P are set. By searching from the arrow X direction Esx and the illustrated arrow Y direction Esy, a portion where the transmission image becomes dark for the first time is detected. In the figure, a method for detecting the address A (X1, Y1) of the area A is shown. In this way, the addresses (X, Y) at the four corners of the bill P are detected. The width and length are calculated from the addresses A (X1, Y1), B (X2, Y2), C (X3, Y3), and D (X4, Y4) at the four corners according to the following formulas 1 and 2.
Width: ((Y1-Y3) + (Y2-Y4)) / 2 ... 1
Length: ((X2-X1) + (X4-X3)) / 2 ... 2
The shape information calculated in this way is transmitted to the detection processing device 7.

  FIG. 5 is a diagram for explaining the bill type detection of the banknote P by the ticket type detection device 5. Here, the case where the bill type of the 10 euro banknote P10 is detected will be described as an example. The detection area Ed for designating the bill type detection range of the bill P is set to horizontal (X) 256 pixels and vertical (Y) 256 pixels. When the bill P is conveyed in the direction of the arrow A (see FIG. 1), the reflected image in the set bill type detection area Ed is read. The resolution per pixel is set to a value suitable for ticket type detection depending on the optical magnification of the ticket type detection device. Note that the ticket type detection area Ed is larger than the other detection areas such as horizontal 1024 pixels and vertical 1024 pixels when the banknote size of the banknote differs depending on the banknote type, and the banknote type detection of the banknote of the maximum banknote length is performed. Because it is necessary, it is set large. In the illustrated 10 euro banknote P10, the U-shaped and face value inverted upside down are printed on the right side of the banknote, and the left side of the banknote is a watermark.

  In the ticket type detection processing unit 5C, a standard pattern generated from a standard image based on a genuine note is registered as a dictionary for each ticket type, front and back, and forward and reverse. The ticket type detection processing unit 5 </ b> C performs a comparison determination between the standard pattern and a pattern generated from the front surface reflection image and the back surface reflection image of the banknote P to be detected by a pattern matching method or the like. This determination result is transmitted to the detection processing device 7.

FIG. 6 is a diagram for explaining authenticity detection of the banknote P by the authenticity detection device 6. Here, a case where true / false of a 10 euro banknote (hereinafter referred to as banknote P10) is detected will be described as an example. The authenticity area Ej1 for designating the authenticity detection range of the banknote P10 is set to horizontal (X) 256 pixels and vertical (Y) 256 pixels. In the case of the banknote P10, the true / false area Ej1 A fake detection area Ej2 (area surrounded by a broken line) is set. When the banknote P10 is conveyed in the direction of the arrow A (see FIG. 1), the reflected image of the surface of the banknote P10 is IR (infrared), R (red), G (green), B in the authenticity detection area Ej2. An image generated by the reflected light of the (blue) component is read. Note that the resolution per pixel is set to a value suitable for authenticity detection depending on the optical magnification of the authenticity detection device. The total value (integrated value) D of the image density values read for each component in this way is calculated by Equation 3, and the feature values DIR, DR, DG, and DB of each component are obtained.

  The total value D of the density values is calculated for the reflection image on the back side of the banknote P10 as well as the front side of the banknote P10, but the description is omitted because it is the same as that on the front side of the banknote P10. The total value D calculated in this way is transmitted to the detection processing device 7.

  FIG. 7 is an example of a discrimination table for changing the discrimination level for discriminating the bill P by the bill discriminating apparatus 100 of the present invention depending on the issuing place of the bill P. FIG. 7A shows a discrimination table in the case of the banknote P10.

  In the true / false detection of banknotes, there is a problem of subtle color differences between the printing office and issuing office that issue banknotes. For this reason, a memory (authentication determination level storage means) 7b is provided which stores in advance the authenticity determination level for detecting authenticity for each ticket type, for each transport state, and for each issuing place. Based on the issuing place corresponding to the first English letter Pb of the serial number Pa read by the device 3, the ticket type detected by the above-mentioned ticket type detection device 5, and the state of conveyance, it is true from the memory (authentication determination level storage means) 7b. A false discrimination level is selected.

  For example, in the case of an Italian-issued banknote with the first alphabetic character Pb of the serial number Pa being “S”, the discrimination table TS10 surrounded by a broken line is selected. For example, the discrimination table 6 is selected when the bill type of the bill to be transported and the transport state are 10 euros / front / positive. Similarly, in the case of 10 euros / table / reverse, the discrimination table 22 is selected. Hereinafter, similarly, the discrimination table 38 is selected for 10 euros, reverse, and normal, and the discrimination table 54 is selected for 10 euros, reverse, and reverse. Similarly, when the first letter Pb of the serial number Pa is “T”, the discrimination table TT10 surrounded by a broken line is selected. The same applies to other cases.

  FIG. 7B shows the authenticity determination level set by selecting the determination table 6. The figure also shows a case where the determination table 5 is selected when the first English letter Pb is “T”. In this way, the discrimination table can be selected based on the ticket type, the transport state, and the serial number, and the authenticity discrimination level is set based on the selected discrimination table. A method for comparing and determining the feature amounts of the respective components and the determination level will be described later.

  FIG. 7 (3) shows the reference ticket size for shape detection set by selecting the discrimination table 6. Since the ticket size may vary depending on the printing station or issuing station, the discrimination table can be selected. However, the ticket size discrimination level does not change depending on the transport state in which the bill is transported. 6, 22, 38 and 54 have common values. Details regarding the ticket size determination level will be described later.

FIG. 8 is a flowchart for explaining the bill discriminating process of the bill discriminating apparatus 100.
The banknote P is taken out from the banknote supply unit 2 (S1). When the taken-out banknote P is conveyed and detected by the timing sensor 10, the detection area of each detection device is set using the detected timing as a trigger (S2).

  Next, OCR processing for reading the serial number of the bill P conveyed by the OCR detection device 3 is performed. The processing result by this OCR processing is transmitted to the detection processing device 7.

  Next, the shape of the banknote P is detected by the shape detection device 4 (S4). The shape information of the bill P measured by the shape detection device 4 is transmitted to the detection processing device 7.

  Next, the ticket type / conveyance state of the banknote P is detected by the ticket type detection device 5 (S5). The ticket type and the transport state detected by the ticket type detection device 5 are transmitted to the detection processing device 7.

  Next, the authenticity feature amount of the banknote P is calculated by the authenticity detection device 6. The authenticity feature amount calculated by the authenticity detection device is transmitted to the detection processing device 7 (S6).

  Next, the banknote detection determination process is performed by the detection processing device 7 based on the OCR processing result, the shape detection result, the bill type detection result, and the authenticity detection result (S7). The authenticity of the banknote is determined by this detection determination process. This determination result is transmitted to the sorting and accumulating device 8.

  Next, the sorting and stacking device 8 sorts and stacks the banknotes P that are transported based on the determination result received from the detection processing device 7 (S8). At this time, the banknote P determined to be a genuine note is stacked in the plurality of stacking units 8c to 8h provided in the sorting and stacking device 8 for each ticket type, or for each ticket type and transport state. Moreover, the banknote P determined to be a fake ticket is accumulated in the rejected ticket accumulating unit 8 b provided in the sorting and accumulating device 8.

  FIG. 9 is a flowchart for explaining authenticity determination processing performed by the detection processing device 7. Here, the case where the banknote P10 is discriminated as described above will be described.

  First, the vertical size and the horizontal size of the bill P measured by the shape detection device 4 are read from the memory (RAM) 7b in which data is stored (S10).

  If the read data exceeds the measurement range and is too long or too short, the shape detection abnormality is set and the process is terminated (NO in S11). On the other hand, if the read data is normal (YES in S11), the ticket type data and the transport state data detected by the ticket type detection device 5 are read from the memory (RAM) 7b in which the data is stored (S12).

  If the ticket type and the transport state are not set in the read data, the ticket type detection abnormality is set because the ticket type detection is abnormal, and the process is terminated (NO in S13). On the other hand, if the read data is normal (YES in S13), the serial number data read by the OCR device 3 is read from the memory (RAM) 7b in which the data is stored (S14).

  If the serial number is not set in the read data, it is an OCR error, so an OCR error is set and the process is terminated (NO in S15). On the other hand, if the read data is normal (YES in S15), the discrimination table is selected based on the detected ticket type, transport state, and serial number. For example, in the case of 10 euros / table / positive issued by Italy, the discrimination table TS10 shown in FIGS. 7 (1) and (2) is selected.

Next, the feature amounts DIR, DR, DG, and DB for each component calculated by the authenticity detection device 6 are read out and compared with the authenticity determination level of FIG.
That is, it is determined whether or not the feature amount DIR falls within the range between the lower limit value 15 of the feature amount and the upper limit value 70 of the feature amount (S17). As a result of this comparison determination, if it is included in the range (YES in S17), the process proceeds to the next processing step. If it is not included in the range (NO in S17), it is determined as a fake note (S24).

  Next, it is determined whether or not the feature amount DR falls within the range between the lower limit value 20 of the feature amount and the upper limit value 80 of the feature amount (S18). As a result of this comparison determination, if it is included in the range (YES in S18), the process proceeds to the next processing step, but if it is not included in the range (NO in S18), it is determined as a fake note (S24).

  Next, it is determined whether or not the feature amount DG is included in the range between the lower limit value 20 of the feature amount and the upper limit value 80 of the feature amount (S19). As a result of the comparison determination, if it is included in the range (YES in S19), the process proceeds to the next processing step. If it is not included in the range (NO in S19), it is determined as a fake note (S24).

  Next, it is determined whether or not the feature amount DB is included in the range between the lower limit value 20 of the feature amount and the upper limit value 80 of the feature amount (S20). As a result of this comparison determination, if it is included in the range (YES in S20), the process proceeds to the next processing step, but if it is not included in the range (NO in S20), it is determined as a fake note (S24).

Next, the vertical size and the horizontal size of the banknote P10 measured by the shape detection device 4 are compared and discriminated with a bill size discrimination level set to ± 2 based on the standard bill size in FIG.
Whether the vertical length (vertical size) falls within the range of the vertical length lower limit value 65 and the vertical length upper limit value 69 is compared (S21). As a result of this comparison determination, if it is included in the range (YES in S21), the process proceeds to the next processing step, but if it is not included in the range (NO in S21), it is determined as a fake note (S24).

  Next, it is compared and discriminated whether the horizontal length (horizontal size) falls within the range of the horizontal length lower limit value 125 and the horizontal length upper limit value 129 (S22). As a result of this comparison determination, if the result is included in the range (YES in S22), the authenticity determination result is determined to be a genuine note (S23), and this authenticity determination process is terminated, but if not included in the range (S22). NO) is determined to be a fake ticket (S24).

  As described above, according to the present embodiment, since the authenticity determination level can be set for each serial number that identifies the issuing country (issuing authority), it is a banknote having a genuine note distribution that differs depending on the issuing authority. However, it is possible to provide a banknote discriminating device provided with a true / false detection device that does not cause performance degradation.

  FIG. 10 is a schematic front sectional view of the bill discriminating apparatus 100 according to the second embodiment of the present invention. FIG. 11 is a block diagram illustrating a configuration of the bill discriminating apparatus 100 according to the second embodiment. Hereinafter, the configuration of the bill discriminating apparatus 100 will be described with reference to FIGS. 10 and 11. In the second embodiment, the authenticity detection device 6 of the bill discriminating apparatus 100 according to the first embodiment shown in FIGS. 1 and 2 is changed to a damage detection device (loss detection means) 9. Since this is the same as that of FIG. 1, the same parts are denoted by the same reference numerals, the description thereof will be omitted, and only the changed parts will be described.

  The damage detection device 9 illuminates the upper surface of the bill P being conveyed and detects the reflected light by the CCD sensor, illuminates the lower surface of the bill P being conveyed, and the reflected light is detected by the sensor. It comprises a lower reflection optical system 9b for detecting and a loss detection processing unit 9c. The upper reflection optical system 9a and the lower reflection optical system 9b read an image in a predetermined damage detection area (details will be described later) set for detecting the damage of the banknote P, and respectively detect the damage detection processing. To the unit 9c. Note that, as described in the setting of the true / false detection area of the first embodiment, the fitness detection area is set for each ticket type because the ticket size (vertical size, horizontal size) may differ depending on the ticket type. It is set for each conveyance state.

  FIG. 12 is a diagram for explaining the detection of the correctness of the banknote P by the correctness detecting device 9. Here, the case where the damage of the 10 euro banknote P10 is detected will be described as an example. When the bill P10 is conveyed in the direction of the arrow A (see FIG. 10), the damage detection area Eu2 (area surrounded by a broken line) in the detection area Eu1 of horizontal (X) 256 pixels and vertical (Y) 256 pixels ) Is read. Each of the read images is output to the damage detection processing unit 9c.

  It should be noted that the resolution per pixel of the read image is set to a value suitable for detection of loss depending on the optical magnification of the loss detection device. In the detection area, an arbitrary pixel value can be referred to using an address (X, Y).

The damage detection processing unit 9c calculates the damage feature amount M of the banknote P using Equation 4 based on the images output from the upper reflection optical system 9a and the lower reflection optical system 9b. The loss characteristic amount M calculated here is a value obtained by squaring the difference between pixels adjacent in the horizontal direction (X direction) in the loss detection area Eu2 and adding all the regions in the loss area Eu2. . Here, the address of an arbitrary pixel in the damage detection area Eu2 is expressed as (X _i , Y _j ), and it is assumed that n pixels are included in the X direction and the Y direction, respectively.

  FIG. 13 is a flowchart for explaining banknote discrimination processing of the banknote discrimination apparatus 100 according to the second embodiment. In this flowchart, the authenticity detection (step S6) in the flowchart for explaining the banknote determination process of the banknote determination apparatus 100 according to the first embodiment shown in FIG. 8 is changed to the correctness detection (step S35). 8 is the same as FIG. 8, the same reference numerals are assigned to the same parts, and the description thereof is omitted. Only the changed parts will be described.

  The damage feature amount M is calculated from the image in the damage detection area Eu2 of the banknote P read by the damage detection device 9 (S35). The calculated damage feature amount M is transmitted to the detection processing device 7.

  Next, a detection determination process is performed by the detection processing device 7 based on the OCR processing result, the shape detection result, the ticket type detection result, and the damage detection result (S36). This detection determination process determines the correctness of the banknote. This determination result is transmitted to the sorting and accumulating device 8.

  Next, the sorting and stacking device 8 sorts and stacks the banknotes P that are transported based on the determination result received from the detection processing device 7 (S37). At this time, the banknote P determined to be a genuine note is stacked on the stacking unit set in any of the stacking units 8c to 8h of the sorting and stacking device 8, and the banknote P determined to be a banknote is sorted and stacked similarly. It is accumulated in the accumulation unit set in any of the devices 8b to 8g. Are accumulated in one of the plurality of stacking units 8c to 8h. Moreover, the banknote P determined to be a fake ticket is accumulated in the rejected ticket accumulating unit 8 b provided in the sorting and accumulating device 8.

  FIG. 14 is a flowchart for explaining the damage determination process performed by the detection processing device 7. Here, the case where the banknote P10 is discriminated as described above will be described. Here, the processes from step S40 to step S46 are the same as steps S10 to S16 of the authenticity determination process of the first embodiment shown in FIG. 9. As a result of this process, the ticket type, the transport state, and the serial number are the same. Detected. Based on the detection result, the damage determination level is selected from the memory (loss determination level storage means) 7b stored in advance for each ticket type, each transport state, and each issuing place. An upper limit is set.

  For example, in the case of an Italian-issued banknote with the first alphabetic character Pb of serial number Pa being “S”, the discrimination table TS10 surrounded by a broken line in FIG. 7A is selected. For example, the discrimination table 6 is selected when the bill type of the bill to be transported and the transport state are 10 euros / front / positive. By selecting this discrimination table 6, an upper limit value of the damage feature quantity M (not shown) is set.

  Next, the damage feature amount M calculated by the damage detection device 9 is read out. A comparison is made to determine whether the read characteristic value M is less than the maximum characteristic value M 30. As a result of the comparison determination, if it is less than the upper limit value (YES in S47), the process proceeds to the next processing step.

Next, the vertical size and the horizontal size of the banknote P10 measured by the shape detection device 4 are compared and determined.
A comparison is made to determine whether the vertical length (vertical size) falls within the range of the vertical length lower limit value 65 and the vertical length upper limit value 69 (S48). As a result of the comparison determination, if it is included in the range (YES in S48), the process proceeds to the next processing step, but if it is not included in the range (NO in S48), it is determined as a damaged ticket (S51).

  Next, it is determined whether or not the horizontal length falls within the range of the horizontal length lower limit value 125 and the horizontal length upper limit value 129 (S49). As a result of this comparison determination, if it is included in the range (YES in S49), the damage determination result is determined to be a genuine note (S50), and this damage determination processing is terminated, but is not included in the range (S50). NO) is determined as a non-performing ticket (S51).

  As described above, according to the present embodiment, since the damage determination level can be set for each serial number that identifies the issuing country (issuing office), it is a banknote having a genuine note distribution that differs depending on the issuing office. However, it is possible to provide a banknote discriminating device provided with a damage detection device that does not cause performance degradation.

  In addition, although the said Example demonstrated the case where a banknote was discriminate | determined, it is not restricted to this, For example, the case of securities, such as a check and a gift certificate, is applicable similarly.

BRIEF DESCRIPTION OF THE DRAWINGS The schematic front view of the banknote discrimination apparatus by Example 1 of this invention. The block diagram which shows the structure of the banknote discrimination apparatus of Example 1. FIG. The figure explaining reading of the serial number of a bill by an OCR device. The figure explaining the shape detection of the banknote by a shape detection apparatus. The figure explaining the bill type detection of the banknote by a ticket type detection apparatus. The figure explaining the authenticity detection of the banknote by an authenticity detection apparatus. An example of the discrimination | determination table for discriminating a banknote with a banknote discrimination | determination apparatus. The banknote discrimination | determination process flowchart of the banknote discrimination apparatus by Example 1. FIG. 5 is a flowchart of authenticity determination processing of the detection processing device according to the first embodiment. The schematic front view of the banknote discrimination apparatus by Example 2 of this invention. The block diagram which shows the structure of the banknote discrimination | determination apparatus by Example 2. FIG. The figure explaining the damage detection of the banknote by Example 2. FIG. The banknote discrimination | determination process flowchart of the banknote discrimination apparatus by Example 2. FIG. FIG. 9 is a flow chart of the damage determination process of the detection processing apparatus according to the second embodiment.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Conveyance path 2 Supply part 3 OCR apparatus 4 Shape detection apparatus 5 Ticket type detection apparatus 6 Authenticity detection apparatus 7 Detection processing apparatus 8 Classification accumulation apparatus 9 Damage detection apparatus 10 Timing detection apparatus P Bill Pa Serial number Pb First of serial number English letter Eo Character reading area Ed Ticket type detection area Ej2 Authenticity detection area Eu2 Damage detection area

Claims (4)

Authenticity determination level storage means for storing the authenticity determination level for determining the authenticity of securities for each type, for each transport state, and for each issue place;
OCR means for reading a symbol indicating an issuing place printed on the securities;
Type detection means for detecting the type and transport state of the securities;
Authenticity information acquisition means for acquiring authenticity information for performing authenticity determination of the securities;
A selection means for selecting the authenticity determination level from the authenticity determination level storage means based on the issuing place corresponding to the symbol read by the OCR means, and the type and transport state detected by the type detection means;
Authenticity determination means for determining authenticity of the securities by comparing the authenticity information acquired by the authenticity information acquisition means with the authenticity determination level selected by the selection means;
A securities discriminating apparatus characterized by comprising: The authenticity information acquisition means includes
A true / false detection area is set based on the type and conveyance state of the securities detected by the type detection means, and the total value of the density values of the reflected images of the securities in the set authenticity detection area A concentration value calculating means for calculating
The authenticity determination means includes
2. The securities determination according to claim 1, wherein a total value of the concentration values calculated by the concentration value calculating means is compared with the authenticity determination level to determine whether the securities are genuine or counterfeit. apparatus. A damage determination level storage means for storing a damage determination level for determining the damage of securities in advance for each type, for each conveyance state, and for each issue place;
OCR means for reading a symbol indicating an issuing place printed on the securities;
Type detection means for detecting the type and transport state of the securities;
A damage information acquisition means for acquiring damage information for determining the damage of the securities;
A selection means for selecting a damage determination level from the damage determination level storage means based on the issuing place corresponding to the symbol read by the OCR means, the type detected by the type detection means and the transport state;
A damage determination means for determining the damage of the securities by comparing the damage information acquired by the damage information acquisition means with the damage determination level selected by the selection means;
A securities discriminating apparatus characterized by comprising: The damage information acquisition means includes
A damage detection area is set based on the type and transport state of the securities detected by the type detection means, and between adjacent pixels of the reflected image of the securities in the set damage detection area. A loss feature amount calculating means for calculating a loss feature amount by adding the difference is provided,
The fitness judgment means includes
4. The securities according to claim 3, wherein the loss feature amount calculated by the loss feature amount calculating means is compared with the loss determination level, and the securities are determined to be a genuine note or a loss ticket. Discriminator.

Images