KR101165497B1 - Apparatus and method for discriminating paper sheet - Google Patents

Abstract

An object of the present invention is to increase the discrimination accuracy of paper sheets in the paper sheet discriminating device.

The determination unit 14 performs pattern matching between the image data of the banknote detected by the image sensor 13 and the data of the surface, the back surface, and the up and down reference of each type of currency stored in the reference data storage unit 15, Determine the currency type and state of the bill. The discriminating unit 16 specifies the optimum detection position of the feature portion of the banknote based on the determination result, and moves the magnetic sensor 17 to that position to detect the feature portion.

Paper sheet discrimination apparatus, judgment part, image sensor, memory part, magnetic sensor

Description

Paper-Paper Discrimination Device and Paper-Paper Discrimination Method {APPARATUS AND METHOD FOR DISCRIMINATING PAPER SHEET}

The present invention relates to an apparatus for discriminating paper sheets, such as banknotes and securities, and a discrimination method thereof.

A banknote discrimination apparatus for discriminating the authenticity of banknotes is mounted on a cash machine of a bank or the like. The banknote discrimination apparatus includes an image sensor for acquiring the image data of the banknote, a magnetic sensor for detecting the magnetic pattern of the banknote, the magnetic data acquired by the sensors, the image data for authenticity judgment and the magnetic data stored in advance. The authenticity of banknotes is discriminated by comparing.

Patent Document 1 describes that the paper sheet identification sensor can be moved in a direction orthogonal to the insertion direction of the paper sheets, and the identification range is expanded by moving the sensor in the zigzag trajectory.

Patent document 2 describes that width position detection means for detecting the position of both ends of a banknote is provided on both sides of the conveying surface of the banknote inlet, and the discrimination accuracy is improved by moving the readhead in accordance with the width of the banknote.

Patent Document 3 describes that the vertical gap between a pair of sensors disposed to face each other is made smaller than the vertical gap between paper conveying paths to reduce the error of the sensor due to disturbance light.

Patent document 4 describes the banknote identifier which made the sensor part which detects banknote identification data movable, and which changed the identification data detection position with respect to banknote.

6 is a flowchart of the discrimination processing of the conventional banknote discrimination apparatus.

The process of the conventional banknote discrimination apparatus is demonstrated using the flowchart of FIG.

First, the image data of the banknote put into the banknote discrimination apparatus is acquired by an image sensor (image sensor process: S21).

Next, magnetic data of the banknote is acquired by the magnetic sensor (magnetic sensor processing: S22).

Next, image data of the banknote irradiated with ultraviolet rays is acquired (UV sensor process: S23).

Then, the data acquired in the processing of each step is compared with the respective data for authenticity determination stored in advance, and it is determined whether the determination result of each sensor is normal (true winding).

When it is determined that the determination result of all the sensor processes is normal (true winding) (S24, YES), the flow advances to step S25 to accept the inserted banknote as a normal banknote. In addition, when there is any one judged abnormal as a result of the sensor processing (S24, NO), the flow advances to S26, and the inserted banknote is determined to be inadequate and rejected.

In the banknote discrimination apparatus which discriminates a plurality of banknotes of different sizes, the width of the conveying path is designed to be larger than the maximum dimension of the banknote used. In addition, the banknote may have a different recording position of a feature where magnetic patterns or the like for authenticity determination are recorded depending on the type of money. For this reason, it is required to be able to accurately discriminate a plurality of types of bills.

Invention of patent document 1 needs to detect a fixed range by reciprocating an identification sensor, and there exists a problem that a discrimination time becomes long.

Although the invention of patent document 2 makes a determination by moving the position of a sensor, the objective is to make the countermeasure easy at the time of a fake banknote generation, and for the purpose of discriminating the authenticity of the several currency types from which the characteristic part of a banknote differs. It is not one.

Conventionally, each sensor (magnetic, UV, etc.) is often mounted in the apparatus one by one in order to suppress component costs of the bill discrimination apparatus. Since the position to be detected shifts due to meandering or the like, it is difficult to accurately detect the features of all banknotes.

[Patent Document 1] Japanese Patent Application Laid-open No. Hei 5-62044

[Patent Document 2] Japanese Patent Application Laid-open No. 58-121491

[Patent Document 3] Japanese Patent Application Laid-Open No. 2003-317130

[Patent Document 4] Japanese Patent Application Laid-open No. Hei 10-31775

An object of the present invention is to increase the discrimination accuracy of paper sheets by accurately acquiring data of a feature of a bill without increasing each sensor.

The paper sheet discriminating device of the present invention is a device for discriminating a plurality of types of paper sheets, an image sensor for acquiring image data of the plurality of paper sheets, identification data for identifying the types of the plurality of paper sheets, and the plurality of paper sheets. A storage means for storing reference data for discriminating paper sheets of the paper, judgment means for comparing the image data acquired by the image sensor with the identification data stored in the storage means, and determining the type of the paper sheets; On the basis of the feature detecting sensor arranged behind the conveyance path of the said image sensor on the conveyance path which conveys the said paper sheet, and movable in the direction orthogonal to a conveyance direction, and the kind of said paper sheet determined by the said determination means. After moving the feature detection sensor to a predetermined position, the paper sheet acquired by the feature detection sensor And a discriminating means for comparing the detected data with the reference data stored in the storage means to discriminate the authenticity of the paper sheets.

According to the present invention, the discrimination accuracy of paper sheets can be improved by changing the detection position of the sensor in accordance with the type of paper sheets.

In the paper sheet discriminating device, the feature detecting sensor has at least a magnetic sensor, and the memory means corresponds to a type and a front and back of the paper sheet to detect the recording position of the magnetic pattern of the paper sheet. Storing the characteristic information, the discriminating means acquires the detection characteristic information according to the type and front and back of paper sheets determined by the determining means, moves the magnetic sensor to the position indicated by the acquired detection characteristic information, and The authenticity of the banknote is discriminated by comparing the detected data with the magnetic pattern data of the corresponding currency type.

With this arrangement, the magnetic pattern can be detected at the optimum detection position according to the currency type and the front and back of the bill, so that the discrimination accuracy of the bill can be improved.

According to the present invention, it is possible to increase the discrimination accuracy of paper sheets.

EMBODIMENT OF THE INVENTION Hereinafter, preferred embodiment of this invention is described. FIG. 1: is a figure which shows the structure of the principal part of the banknote discrimination apparatus (paper sheet discrimination apparatus) 11. As shown in FIG.

When a banknote is put into the input part 12, a banknote is conveyed by the conveyance path which is not shown in figure, and the image data of a banknote is read by the image sensor 13 initially. The determination part 14 performs rotation processing etc. with respect to the image data of the banknote detected by the image sensor 13, aligns image data in a fixed direction, and the reference | standard of the banknote stored in the reference data storage part 15 The money type, front and back, and top and bottom of the banknote are determined in comparison with the image data to be used. This determination result is output to the discriminating unit 16.

The reference data storage section 15 includes image data (data serving as a standard for determining the types of paper sheets) of the upper and lower banknotes, the surface of each currency type, the back surface, and a detection indicating the data and the position of the feature portion of the bill. Characteristic information (data serving as a standard for discriminating paper sheets) is stored.

The discrimination unit 16 obtains, from the reference data storage unit 15, the detection characteristic information indicating the position of the characteristic portion of the corresponding currency type, from the reference data storage unit 15, when the determination result of the currency type, the up and down, and the front and back of the bill is obtained. The position of the magnetic sensor 17 is moved based on the information.

The magnetic sensor 17 is arrange | positioned at the position of the back of the image sensor 13 on a conveyance path, and is attached to the actuator 18 (refer FIG. 4) so that a movement is possible. The magnetic sensor 17 is movable in the direction orthogonal to the conveyance path, and moves to the position indicated by the discriminating unit 16 to detect magnetic data of the banknote.

The discriminating unit 16 compares the magnetic data detected by the magnetic sensor 17 with the magnetic pattern data of the corresponding currency type stored in the reference data storage unit 15 and determines whether the two coincide with each other. The authenticity of the bill is determined.

The determination unit 14 and the discriminating unit 16 described above can be realized by, for example, a CPU and a discrimination processing program described later. The determination unit 14 and the discriminating unit 16 may be configured by one module.

2 is a flowchart of the discrimination processing of the banknote discrimination apparatus 11 according to the embodiment.

Initially, image sensor processing is performed on a banknote (S11). In this image sensor process, for example, the distance ΔX (see FIG. 4) from the position where the banknote first passes the image sensor 13 to the end of the conveying path and the meandering angle of the banknote are calculated to correct the inclination of the banknote. Convert to image data.

Next, pattern matching processing is executed (S12). In this pattern matching process, pattern matching between the converted image data and data serving as a reference for the bills of a plurality of currency types stored in the reference data storage unit 15 is performed to identify a currency type having the highest matching rate.

Next, with reference to the banknote information table 21 memorize | stored in the reference data storage part 15, detection characteristic information which shows the position in which the magnetic pattern of the determined currency type is recorded is acquired (S13).

3 is a diagram illustrating the configuration of the banknote information table 21. In the banknote information table 21, the detection characteristic information indicating the position where the characteristic information (magnetic pattern, etc.) of the banknote is recorded in correspondence with the state (upward, downward, front, back, etc.) of the banknote for each currency type is stored. It is.

In the banknote information table 21, for example, as shown in Fig. 3, the surface of the A currency type, the recording position of the characteristic information (magnetic pattern, etc.) in the upward state is 120 mm, the surface of the B currency type. It is stored as detection characteristic information that the recording position of the characteristic information of the downward state is 65 mm. Therefore, when the currency type, front and back, and top and bottom of a banknote are specified, the detection characteristic information of the corresponding banknote can be read from the banknote information table 21.

Next, the flow advances to step S14 of FIG. 2 to execute an actuator control process. In this actuator control process, the actuator 18 is controlled based on the detection characteristic information read out from the banknote information table 21, and the magnetic sensor 17 is moved to an optimal position for detecting the feature of the banknote.

Next, a matching process of the bill feature is executed (S15). In this matching process, the information (magnetic pattern data) of the feature of the corresponding currency type is stored in the reference data storage section 15, and the information of the read feature and the detection data of the magnetic sensor 17 are read. Pattern matching is performed.

In the present embodiment, the magnetic sensor 17 is moved to an optimal position for detecting the feature of the bill according to the currency type, front and back, and top and bottom states of the bill to be discriminated. Thereby, since the magnetic sensor 17 can always detect the feature part of a banknote, it is not necessary to store the data used as the reference | standard of pattern matching for the area | region of the whole banknote, and only the data of a characteristic part need only be memorize | stored in a memory. Specifically, since only the data (this is called a template) of the portion in which characteristic information such as a magnetic pattern is recorded for one currency type is stored in the memory, it is possible to reduce the amount of data to be stored in the memory.

Next, it is determined whether or not the result of the entire sensor processing is normal (S16). If the detection data of the image sensor 13 coincides with the image data of the specific banknote, and the detection data of the magnetic sensor 17 coincides with the magnetic pattern data of the corresponding banknote (S16, YES), step Proceed to S17 and accept as a normal banknote. On the other hand, if any of the sensor's processing result is NG (S16, NO), the flow advances to step S18, where it is determined as a fake banknote and the banknote is rejected.

4 is a diagram illustrating a relationship between the feature portion of the banknote and the detection position of the magnetic sensor 17.

The width | variety of the conveyance path of the banknote discrimination apparatus 11 is designed larger than the largest dimension of the banknote 22 in the longitudinal direction. Therefore, the banknote 22 put in is often conveyed in the oblique state. For example, the meandering angle θ of the banknote can be calculated from the time difference between the banknote passing through any two points (A point, B point) of the image sensor 13 and the distance L between the two points and the conveying speed. Moreover, the distance (DELTA) X which referred to the side surface of a conveyance path can be calculated from the position which the right side of the banknote 22 initially passed the image sensor 13.

When the image data of the whole banknote 22 is detected by the image sensor 13, the determination part 14 correct | amends the meandering angle of the detected image data, and converts it into the image data which makes a specific position the origin (normalizes). . Then, pattern matching is performed with the reference data in the up-and-down direction on the surface and back surface of each currency type stored in the reference data storage section 15, and it is determined that the banknote 22 is the currency type having the highest coincidence rate.

The discrimination unit 16 detects the characteristic information indicating the optimal detection position of the feature 22a of the banknote 22, if the difference between the money type, the surface, the back side, and the top and bottom of the banknote 22 is determined by the determination unit 14. Is obtained from the banknote information table 21. Based on the acquired information, the actuator 18 is driven to move the magnetic sensor 17. The amount of movement of the magnetic sensor is calculated from the banknote information table 21 and the distance ΔX from the side surface of the conveying path.

As a specific operation procedure, as shown in FIG. 4, the position of the present magnetic sensor is compared with the positional information by the detection characteristic information obtained from the banknote information table, and it is determined whether the magnetic sensor should be moved. And when it is determined that it should be moved, the movement of the magnetic sensor is completed until the banknote is conveyed to the position of the magnetic sensor. And the magnetic part 17 detects the characteristic part 22a 'of the banknote 22. FIG. Thereby, a different feature part can be detected correctly according to the kind of money, the surface, the back surface, and the up-down state of the banknote 22.

FIG. 5 is a diagram showing the configuration of the feature portion 22b of the banknote 22 and the template. FIG. 5A shows an example in the case where the feature portion (the portion on which the magnetic pattern is recorded) 22b is present in the portion on the right side of the banknote 22.

When the currency type, the back surface, the surface, and the top and bottom of the banknote 22 are specified by the image data detected by the image sensor 13, the detection characteristic in which the feature of the corresponding banknote 22 exists from the banknote information table 21 is present. Information is obtained. And based on the detection characteristic information, since the magnetic sensor 17 is an initial position, it moves to the right direction from the center of the actuator 18, and detection of the characteristic part of a banknote is performed. At this time, if the banknote 22 meanders, the feature will also be detected in an oblique direction with respect to the banknote 22.

Therefore, the data of the template to be read from the reference data storage section 15 also read data by the trajectory (trace represented by the dotted line in Fig. 5) according to the meandering angle of the banknote, and the read data and the magnetic sensor 17 Pattern matching of the detection data is performed.

By moving the magnetic sensor 17 in the direction orthogonal to the conveying path as described above, the data of the characteristic portion of the bill can be detected at all times, so that the data of the template serving as a reference for pattern matching can be established. It is sufficient to store only the data of the feature of the city in the memory. This makes it possible to reduce the amount of data of the template to be stored in the memory.

According to the above-described embodiment, the image sensor 13 specifies the currency type, the surface, the back side, and the top and bottom of the banknote, acquires the detection characteristic information indicating the position of the feature portion of the banknote according to the specific condition, and the detection characteristic information. By moving the position of the magnetic sensor 17 on the basis of, the detection can be performed at an optimum position according to the type of currency of the bill. Thereby, since the characteristic part of all kinds of banknotes can be detected correctly, the discrimination precision of the authenticity of banknotes can be improved.

(1) The present invention is not limited to banknotes, but can be applied to paper sheets including documents, such as securities that require authenticity determination.

(2) The sensor for detecting the characteristic information is not limited to the magnetic sensor but may be another sensor such as an optical sensor.

BRIEF DESCRIPTION OF THE DRAWINGS The figure which shows the structure of the banknote discrimination apparatus of embodiment.

2 is a flowchart of the discrimination processing of the embodiment;

3 is a diagram illustrating a configuration of a bill information table.

4 is a diagram illustrating a relationship between a feature portion of a bill and a detection position of a magnetic sensor;

Fig. 5 is a diagram showing the features of a bill and the configuration of a template.

6 is a flowchart of a conventional discrimination process.

<Explanation of symbols for the main parts of the drawings>

11: banknote discrimination device

12: input part

13: image sensor

14: judgment unit

15: reference data storage

16: discrimination

17: magnetic sensor

18: Actuator

Claims (4)

It is a device for discriminating plural kinds of paper sheets, An image sensor for acquiring image data of the plurality of paper sheets; Storage means for storing identification data for identifying types of the plurality of paper sheets and reference data for discriminating the plurality of paper sheets; Determination means for comparing the image data acquired by the image sensor with the identification data stored in the storage means to determine the type of the paper sheet; A feature detecting sensor disposed behind a conveying path of the image sensor on the conveying path for conveying the paper sheets and movable in a direction orthogonal to the conveying direction; Based on the kind of the paper sheet determined by the determining means, the feature detecting sensor is moved to a predetermined position, and then stored in the detected data of the paper sheet acquired by the feature detecting sensor and the storing means. And a discriminating means for discriminating the authenticity of the paper sheets by comparing the reference data. The sensor of claim 1, wherein the feature detection sensor has at least a magnetic sensor, The storage means stores the detection characteristic information indicating the recording position of the magnetic pattern of the paper sheet in correspondence with the type and front and back of the paper sheet, The discriminating means acquires detection characteristic information according to the type and front and back of paper sheets determined by the determining means, moves the magnetic sensor to a position indicated by the acquired detection characteristic information, and detects data of the magnetic sensor, and A paper sheet discriminating device for discriminating authenticity of the paper sheets by comparing magnetic pattern data of a currency type. The method according to claim 1 or 2, wherein the determination means determines the type of money, front and back and top and bottom of the paper sheet, The storage means stores the detection characteristic information indicating the position of the feature portion along the money type, front and back, and top and bottom of the paper sheet, And the discriminating means moves the feature detecting sensor to a position indicated by the detecting feature information. It is a differentiation method to discriminate plural kinds of paper sheets, Image data of the plurality of paper sheets is detected by an image sensor, Identification data for identifying the type of the paper sheets and reference data for discriminating the plurality of paper sheets are stored in the storage means, Comparing the image data of the image sensor with the identification data to identify the type of the paper sheet; The feature detection sensor, which is arranged behind the conveyance path of the image sensor on the conveyance path for conveying the paper sheet and which is movable in a direction orthogonal to the conveying direction, is characterized by the feature of the paper sheet that is to be discriminated according to the type of the paper sheet. To the position to be detected, The paper sheet discriminating method of discriminating the authenticity of the paper sheet by comparing the detection data of the feature detection sensor and the reference data.

Images