JPH0789386B2 - Authenticity determination device for paper sheets - Google Patents

Description

Description: FIELD OF THE INVENTION The present invention relates to a genuine / counterfeit discriminating apparatus for paper sheets, and more particularly, for discriminating the genuine / counterfeit of a bill in a money changer, an automatic teller machine, a vending machine, or the like, or The present invention relates to a genuine / counterfeit discriminating device for paper sheets used for discriminating the authenticity of a check.

[Background of the Invention]

In the authenticity discrimination device for paper sheets, for example, the bill discriminating device,
Generally, as shown in FIG. 8, while the inserted bill 1 is conveyed in the long side direction, the pattern of the bill is read by the sensor 2 fixed on the conveying path and stored as this pattern (hereinafter, referred to as a detection pattern). The pattern comparison circuit 4 compares a pattern (hereinafter referred to as a reference pattern) stored in advance in the circuit 3 to determine the authenticity of the bill 1.

Here, as the reference pattern, in order to increase the probability of matching with the detection pattern, the pattern has a width as shown in FIG. 9, but this is the same as that of a true bill (hereinafter referred to as a true bill). All patterns that can be genuine are included regardless of the distribution frequency of the patterns. Then, the authenticity of the bill is determined by simply comparing whether the detection pattern exists within the reference pattern.

Therefore, in such a conventional bill discriminating apparatus, there is a problem that even a genuine bill is determined to be false if there is a large variation due to a difference in printing density of bills, dirt due to circulation, and the like. In order to increase the probability that the genuine bill detection pattern and the reference pattern match, it is necessary to widen the setting width of the reference pattern.

However, by widening the setting width of the reference pattern, the problem that the genuine note is determined to be false does not occur, but the characteristics of the genuine note on the reference pattern are blurred, and something other than the genuine note, for example, altered or There is a problem that a forged banknote (hereinafter referred to as a counterfeit banknote) is likely to be mistakenly determined to be true. Therefore, in the bill discriminating apparatus, there has been a strong demand for solving the conflicting problems described above.

[Object of the Invention]

The present invention solves the above-mentioned problems in the conventional type, and in the authenticity determination device for paper sheets, the probability of erroneously determining a genuine note as false is reduced without lowering the ability to discriminate against a false note. The purpose is to

[Structure and Effect of Invention]

To achieve the above object, in the present invention, the authenticity of a paper sheet is determined by comparing a detection pattern obtained by detecting a desired position on the paper sheet with a sensor with a reference pattern. A false discriminating device, which determines whether or not the detection pattern is included in the area of the reference pattern for each predetermined section of the desired position on the paper sheet, and predetermined for each section based on the determination result. While giving the evaluation point of, the predetermined calculation is performed on the evaluation point over the entire desired portion to obtain a matching rate that is the degree of coincidence between the detection pattern and the reference pattern, and the matching rate is equal to or greater than a predetermined value. In the true / false discriminating device for determining the paper sheets as true when, the reference pattern is composed of a plurality of basic patterns divided according to the distribution frequency of the pattern of many true paper sheets, Serial evaluation points, characterized in that depending on whether the detection pattern is included in the area of any of the basic pattern.

According to this, more detailed evaluation according to the area width of the reference pattern that is different for each classification is reflected, and it becomes possible to reflect the features such as the pattern and the character of the paper sheet in the matching rate in more detail, and thereby It is possible to accurately capture the characteristics of paper sheets and improve the accuracy of authenticity determination. That is, it is possible to prevent erroneous determination of banknotes due to variations in the printing density of banknotes and variations in circulation stains without lowering the ability to identify counterfeit notes, and to improve processing efficiency.

[Explanation of Examples]

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

Generally, the bill detection pattern is
As shown in Fig. 10, it is widely distributed in the direction of the level axis, but this is a comprehensive distribution, and the distribution frequency of each banknote pattern is The distribution frequency of the part is high, but that of the peripheral part is low.

Focusing on the above point, the present invention provides a reference pattern composed of a plurality of basic patterns for each distribution frequency of banknote patterns, compares the detected pattern with the reference pattern, and from the result, the degree of matching with the reference pattern, that is, matching. Authenticity is determined by obtaining the rate. Specific examples will be shown below.

FIG. 1 shows a hardware configuration of a bill validator according to an embodiment of the present invention.

The apparatus of the figure, the sensor 10 (10 _1), AD conversion circuit 11, microprocessor Seth central processing unit such as a service (CPU) 12, having memory 13 and peripheral INTERFACE (PIA) 14 and the like. DB
Is a data bus, CB is a control bus, and AB is an address bus.

The sensor 10 is fixed at a desired position on a banknote transport path (not shown) and detects magnetism or transmitted or reflected light at a desired scanning line (track) of the banknote. As the sensor 10, a plurality of sensors of the same kind or different kinds such as magnetic and optical for detecting patterns on the same track or different tracks may be arranged. In this case, a plurality of sensors 10 ₁ , ..., 10 _n and the AD conversion circuit 11 are connected via a multiplexer 15, and the outputs of each sensor 10 ₁ , ..., 10 _n are sequentially selected by the multiplexer 15. If the AD conversion circuit 11 is supplied to the AD conversion circuit 11 and the AD conversion circuit 11 is used in a time division manner, one AD conversion circuit 11 is sufficient. SEL is a sensor select signal to the multiplexer 15.

The A / D conversion circuit 11 converts the output DTA of the sensor 10 into digital data DTD at every predetermined sampling time according to the control signal CTR from the CPU 12. This digital data DTD
Is stored in a predetermined area of the memory 13 via the data bus DB under the control of the CPU 12. As a method of storing the digital data DTD of the detection pattern in the memory 13, if a short processing time is required, a direct memory access controller (DMAC) 16 may be provided and a DMA method may be adopted using this DMAC 16. The DMAC 16 sequentially sends the data DTD from the AD conversion circuit 11 to the data bus DB by the DMA request signal DMAR, accesses the memory 13 via the address bus AB, and sends the write command to the control bus CB to send the above data. The DTD is directly stored in the memory 13 without going through the CPU 12. DMA end signal DMA from DMAC16
E is a signal that prohibits the transmission of the data DTD.

In the memory 13, as shown in FIG. 2, in addition to the CPU control area in which the control program of the CPU 12 is stored, basic pattern data a ₁ , a ₂ ,
The reference pattern areas A and B in which a _n and b ₁ , b ₂ , ..., b _n are stored, and the detection pattern data DTD = c ₁ , c ₂ , ... c _n for each sampling time are stored. The detection pattern area C is set.

Next, the determination algorithm of the apparatus of FIG. 1 will be described with reference to the memory table of FIG. 2 and the pattern diagrams and graphs of FIGS.

The pattern data of a large number of genuine bills obtained in advance is overlaid, and the distribution frequency (Fig. 10) is calculated.
A region where% or more is distributed (Fig. 3) and a region where 1% or more and less than 95% are distributed (Fig. 4) are obtained, and these are defined as basic patterns A and B, respectively. A combination of these basic patterns A and B becomes a reference pattern.

The reference patterns A and B obtained above are level data a _{1 to} a _n , b ₁ to b for each unit in the banknote transport direction in the memory.
It is stored as _n (FIG. 2).

Reading ivy detection pattern C is a level data c ₁ to c _n of each segment equal to the reference pattern in the conveying direction of the bill,
It is temporarily stored in the memory (Fig. 2).

The matching rate is the detection pattern C and the reference pattern A,
An evaluation score as shown in the table below is given between B, and this is calculated by comparing the level data 1 to n for each unit division and taking the sum thereof. Then, the one with the larger sum will have a higher matching rate.

Then, the one that satisfies the predetermined matching rate is determined to be true (FIGS. 5 and 6).

In the above example, two patterns A and B are provided as basic patterns, but the distribution frequency is divided into
By providing a plurality of basic patterns, more detailed determination can be performed. Further, the evaluation points can be set in a stepwise manner in accordance with the division of the basic pattern, so that the difference in the matching rate between the genuine bill and the counterfeit bill can be made larger to facilitate the identification.

Furthermore, it is also possible to determine the authenticity by obtaining the matching rate for a partial pattern such as a picture or character of a bill.

Therefore, by the above operations, it is possible to accurately capture the characteristics of the banknote, and even for large variations,
The probability that the detection pattern and the reference pattern match can be increased.

Next, the operation of the apparatus shown in FIG. 1 will be described with reference to the flow chart shown in FIG.

In the standby state, when a banknote is inserted into a banknote insertion slot (not shown), the apparatus of the figure starts the following authenticity determination process.

First, the pattern c ₁ of the bill to be conveyed are inserted in step 1, c _2, ..., reads c _n, the detection pattern c _1, c ₂ in step 2, ..., temporarily stores the c _n in the memory 13 Let Then, in step 3, the basic pattern a _k and the detection pattern c _k are set to k.
= 1 to n are sequentially compared, and the first matching rate a is calculated from the number of data matching the basic pattern A in step 4 and the evaluation point (here, 2 points) when the basic pattern A is matched, and temporarily. Remember. Next, in steps 5 and 6, the second matching rate b is calculated and temporarily stored in the same manner as in steps 3 and 4 for the part in which the evaluation point matches one basic pattern B. In step 7,
The sum of the matching rates a and b is calculated.

Then, in Steps 8 and 9, the sum of the calculated matching rates is compared with a predetermined reference value of the matching rate to make a true / false decision.

There is also a method of comparing the detection pattern c with the reference patterns A and B at the same time as the pattern comparison method of steps 3 to 7 to obtain the sum of the matching rates. According to this method, the number of processing steps is reduced, so that the time required for identification can be shortened.

[Brief description of drawings]

FIG. 1 is a block diagram showing the construction of a bill discriminating apparatus according to an embodiment of the present invention, FIG. 2 is a memory map showing the construction of a memory in the apparatus of FIG. 1, and FIGS. 3 and 4 are of FIG. FIG. 5 is a graph showing areas in which 95% or more and 1% or more and less than 95% of the detection patterns in the apparatus are distributed, and FIG. 5 shows a state in which the reference pattern and the detection pattern are overlapped to obtain an evaluation point for each unit classification. A graph, FIG. 6 is a graph showing an example of the matching rate of each of genuine bills and counterfeit bills obtained by the apparatus of FIG. 1, FIG. 7 is a flow chart for explaining the operation of the apparatus of FIG. 1, and FIG. FIG. 9 is a schematic block diagram of a conventional bill discriminating apparatus, FIG. 9 is a graph showing a region where a detection pattern is true in the apparatus shown in FIG. 8, and FIG. 10 is a variation of a genuine bill detecting pattern in the bill discriminating apparatus. Graph A. 10 …… Pattern identification sensor, 11 …… AD converter, 12 …… CP
U, 13 ... Memory.

Claims (1)

[Claims] 1. An authenticity determination device for a paper sheet for determining the authenticity of the paper sheet by comparing a detection pattern obtained by detecting a desired portion on the paper sheet with a sensor with a reference pattern. Then, it is determined whether or not the detection pattern is included in the area of the reference pattern for each predetermined section of the desired position on the paper sheet, and a predetermined evaluation point is given for each section based on the determination result. At the same time, a predetermined calculation is performed on the evaluation points over the entire desired portion to obtain a matching rate that is the degree of coincidence between the detection pattern and the reference pattern, and the paper sheet when the matling rate is equal to or higher than a predetermined value. In a true / false determination device for a paper sheet that determines a class as true, the reference pattern includes a plurality of basic patterns divided according to the distribution frequency of a large number of genuine paper patterns, and the evaluation point is the detection pattern. Authenticity discrimination device of the sheet, wherein different depending over emissions is included in the area of any of the basic pattern.