JPS5872290A - Discrimination system for paper money or the like - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a method for identifying banknotes, etc., which determines the authenticity of banknotes by reading, for example, patterns drawn on the banknotes.

In this type of conventional discrimination method, standard data regarding banknote patterns is stored in a memory in advance, and the authenticity of banknotes is determined based on the standard data and data read from a detected object. However, in the conventional method, the reference data takes a fixed range of values, so if the sensitivity, performance, etc. of the pattern reading section changes over time, the identification device may malfunction.
This poses a problem in determining the authenticity of banknotes. In addition, if there are variations in characteristics of the structure or component parts of the pattern reading section, etc.,
This results in different classification performance depending on the classification device. For this reason, careful adjustments are required when the device is manufactured, and after it starts operating, it is necessary to periodically perform maintenance and inspection work such as replacing parts and cleaning, which places a heavy burden on those involved.

The present invention automatically corrects variations in the performance of each discrimination device and changes over time by correcting the contents of the standard data based on the read data for appropriateness determination, thereby realizing stable authenticity determination of banknotes. The purpose of this study is to provide a novel identification method.

The present invention will be specifically described below based on embodiments shown in the drawings.

Although the drawings show an embodiment in which the present invention is applied to determine the authenticity of banknote patterns, the present invention is not limited to this, but can also be applied to determine the authenticity of patterns other than banknotes, position, defects, and other various types of determination. .

FIG. 1 shows a situation in which a banknote 1 is being moved along a conveyance path 2 by means of a conveyor bell (20, 20).

This conveyance path 2 may be, for example, a banknote conveyance path inside an automatic teller machine, a currency exchange machine, or the like. Banknote 1 has letters on the front and back sides,
A pattern such as a design is printed, and a specific track 11
The upper pattern is read by a pattern reading device 3 disposed on the conveyance path 2.

This pattern reading device 3 may be of various types such as optical means or magnetic means, but in this embodiment, a photoelectric sensor is provided corresponding to the track 11 of the banknote 1 to detect the amount of transmitted light or reflected light. The pattern for one track is read based on the following. Note that the present invention is not limited to the examples,
For example, it can be applied to a device that reads patterns of multiple tracks.

FIG. 2 shows an example of the circuit configuration of the bill validating device.

In the illustrated example, the arithmetic control unit 4 (hereinafter referred to as 1'-CPUJ) is composed of a microprocessor, decodes an instruction taken out from the memory 5, reads data at an address specified by the instruction from the memory 5, and performs calculations. The results are stored in the memory 5. Further, the CPU 4 transmits the banknote pattern reading signal 11 obtained by the reading device 3 to the interface 6.
controls the operation of reading into the memory 5 via the . The interface 6 converts the analog reading signal fl into a digital pattern data signal f2, and this data signal f2 is divided into a time interval (1, 2) as shown in FIG.
, 3, ..., n') section data in which the signal level changes stepwise ■1°V, , V3.・・・・・・
..., Vn (hereinafter referred to as "read data").

The memory 5 stores various data necessary to determine the authenticity of banknote patterns. FIG. 4 shows the memory table, where the storage area Ml contains a coefficient J1tJz for setting the condition for the permissible range of read data, the storage area M2 contains the number of times N of discriminations that have been judged as "true", and the storage area M3. is the cumulative data value AV for each time interval 1 to n! . . . AVn are respectively stored, and the memory IJA M4 is used as an area for temporarily docking data for correcting the cumulative value in each time period 1 to n.

FIG. 5 shows the interval counter i and the register group ROM, lR?, included in the CPU 4. The contents of each register group RO are shown.
"" R7 contains the number of discriminations N, the determination coefficient J1+J2, the read data Vi of the time interval indicated by the counter value, the data cumulative value AVi, and the data average value aVi.

A maximum data tolerance value aVi (maw) and a minimum data tolerance value aVi (min) are respectively set.

FIG. 6 shows the operation flow of the discrimination method according to the present invention,
The CPU 4 initializes the contents of the counter i to 1 in step 81, and obtains the number of discriminations N and the determination coefficient J1 . Read Jt into registers RO, R1°R2. Then CPU4
reads the read data 1 of time interval 1 obtained from the reading device 3 via the interface 6 and the cumulative value AV1 of time interval 1 stored in storage area M3 into registers R3 and R4, respectively (step 84. 85).

After Δ, the CPU 4 calculates the average value aV1. of the time interval 1. The maximum data permissible value aV l(max) and the minimum data permissible value aV l(m1n) are sequentially calculated using the following formulas, and the results of the calculations are set in registers R5 to R7, respectively (steps 86 to 88).

The maximum data allowable value aV1 (max) and the minimum data allowable value aV1 (min) obtained from the above formula 00 are
The upper and lower limits for determining that the read data (1) is appropriate are shown, respectively. Furthermore, the determination coefficient J1. Jz takes a percentage value (c), and is determined by approximately one experiment depending on the distribution of a plurality of read data.

Thus, in step 89, the read data ■l is equal to the minimum data tolerance value aV1 (min) and the maximum data tolerance value aV.
1 (max), and if the judgment is NO, it is determined that the pattern does not match, and the discrimination operation is immediately stopped, and error processing such as returning the banknote is executed (step 90 ).

If the determination in step 89 is YES, the CPU 4 calculates a revised cumulative value by adding the cumulative value AV1 of time interval 1 and the read data ■1, and temporarily docks this value in the storage area M4 of the memory 5 (step 91 .92).

Next, in step 93, 1 is added to the value of counter i, and the processes of steps 84 to 92 are executed for the read data 2 of the next time interval 2, and the set value of counter i is the final value of the time interval. Similar processing is repeated until n is exceeded, that is, until the determination at step 94 becomes YES. In this repetitive operation, all the read data of time intervals 1 to n are 1 to Vn, which are general 2, etc.
..., n), all read data ■
1 to Vn are appropriate, and the authenticity determination of the banknote 1 is "true". In this case, the storage area M4 stores the revised cumulative value AV of each time interval 1 to n obtained in steps 91 and 92.
1-AVn is set, and in step 95 this data is moved to storage area M3, and in step 96, 1 is added to the number of discriminations N to rewrite the contents of storage area M2. As a result, all of the discrimination operations are completed (step 97), and the authenticity of the next banknote is determined based on the data values corrected above.

As described above, the present invention self-corrects the contents of the reference data for each appraisal process based on the read data for authenticity determination, so even if the sensitivity, performance, etc. of the pattern reading section change over time, The contents of the reference data are sequentially updated, so there is no malfunction of the device, and the discrimination performance can be stably maintained.

In addition, even if there are variations in the characteristics of equipment component parts, the standard data unique to each equipment is set accordingly, so there is no need for careful adjustment during production, reducing the workload of those involved and improving productivity. It has many excellent effects, such as contributing to sexual improvement.

[Brief explanation of drawings]

FIG. 1 is an explanatory diagram showing a banknote discrimination situation, FIG. 2 is a block diagram showing an example of a circuit configuration, FIG. 3 is an explanatory diagram showing a data structure, and FIG. 4 is an explanatory diagram showing a data storage situation in a memory. FIG. 5 is an explanatory diagram showing internal data of the CPU, and FIG. 6 is a flowchart showing the discrimination method according to the present invention. 1...Banknotes 3...Reading device 4...Arithmetic processing device 5...Memory patent applicant Tateishi Electric Co., Ltd.

Claims (1)

[Claims] This method compares data read from a detected object such as a banknote with reference data to determine the suitability of the detected object. A method for identifying banknotes, etc., characterized by correcting data contents.