JP2009238090A - Paper sheet discrimination device and method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a paper sheet discrimination device, wherein a paper sheet with folding at an edges is accepted as the normal one. <P>SOLUTION: When an thickness abnormality is detected by thickness sensor processing (Fig.2, S14, NO), the processing proceeds to a step S16 to determine whether or not density of pixels of a transmitted image at the edge of paper money exceeds a predetermined value. When the thickness of the paper money exceeds an upper limit value and the density of the pixels of the transmitted image at the edge of the paper money exceeds the predetermined value (S16, YES), it is determined that there is the folding of the paper money, the processing proceeds to a step S15 to accept the paper money as a normal paper money. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

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

  A banknote discriminating apparatus for discriminating banknotes is mounted on an automatic teller machine such as a bank. The banknotes inserted into the automatic teller machine may be bent, and these banknotes may be rejected without being judged to be genuine even if they are genuine. Therefore, the ratio of banknotes that cannot be accepted by the automatic teller machine increases, and customer convenience decreases.

  In Patent Document 1, by investigating whether or not there is a sudden transmission change state in which a portion where the light transmittance of the banknote is small and a portion where the light transmittance is large exist, the cause of the thickness abnormality is It is described that it is determined whether or not it is due to smear breakage.

Patent Document 2 describes that a thickness detection sensor and a reflection / transmission sensor are provided, and a corner-folded banknote is determined as a genuine note based on detection results of these sensors.
Patent Document 3 includes a first detection unit that detects the shape through the entire paper sheet, and a second detection unit that detects the thickness of the entire paper sheet. It is described that the judgment of paper sheets is correct and judgment is made by combining the judgments.

  In Patent Document 4, a first slice level for detecting a paper sheet and an overlapping or folding of the paper sheet are detected with respect to a scanning output of an optical scanning unit that optically scans the paper sheet. It is described that an overlapping or broken state can be detected by providing a means for setting the second slice level.

FIG. 4 is a flowchart of the discrimination process of the conventional banknote discrimination device.
The conventional banknote discrimination process is demonstrated using the flowchart of FIG.
First, the image data of the banknote thrown into the banknote discrimination device is acquired by the image sensor. (Image sensor processing: S21).

Next, the magnetic data of a banknote is acquired with a magnetic sensor. (Magnetic sensor processing: S22)
Next, bill thickness data data is acquired by the thickness sensor. (Thickness sensor processing: S23)
Then, the data acquired in the process of each step is compared with each data for authenticity determination stored in advance, and it is determined whether the determination result of each sensor is normal (genuine bill).

  When it is determined that the determination results of all the sensor processes are normal (genuine bills) (S24, YES), the process proceeds to step S25, and the inserted banknote is accepted as a normal banknote. If there is at least one item determined to be abnormal as a result of the determination of the sensor process (S24, NO), the process proceeds to S26, and the inserted banknote is determined to be unacceptable and rejected.

  FIG. 5 is a diagram illustrating a forged banknote and a folded banknote. For example, as shown in FIG. 5A, there is a method in which blank paper is attached to a 5,000-yen bill with a tape so that the outer dimensions are the same as those of a 10,000-yen bill so that it is erroneously recognized as 10,000 bills. Since the thickness of such a banknote bonded with the tape is increased, the counterfeit banknote can be detected by detecting the thickness of the banknote with the thickness sensor.

However, the banknotes put into the automatic teller machine may be bent at the edges of the banknotes, and when all the banknotes whose thickness exceeds the upper limit as in the conventional discrimination process are rejected, FIG. There was a problem that all normal banknotes such as those shown in B) were rejected.
JP-A-4-242833 Japanese Unexamined Patent Publication No. 7-14048 JP-A-9-203628 JP 2000-298101 A

  An object of the present invention is to accept a genuine paper sheet having a fold at the end as a normal one and reliably reject a fake paper sheet that has been bonded in a paper sheet discrimination device. Is to do.

  The paper sheet discrimination apparatus according to the present invention includes a thickness sensor that detects the thickness of the paper sheet, an image sensor that detects a reflection image and a transmission image of the paper sheet, and the thickness sensor determines the thickness of the paper sheet. When it is detected that the upper limit value is exceeded, and further, when the image sensor detects that the pixel density of the transmission image at the edge of the paper sheet exceeds a predetermined value, And a control unit that determines that the end portion is bent and receives the input paper sheet as a normal paper sheet.

According to this paper sheet discrimination apparatus, a paper sheet whose end is bent can be accepted as a normal paper sheet.
The paper sheet discrimination apparatus includes a storage unit that stores a plurality of templates serving as criteria for determining the type of the paper sheet, and the image sensor has a function of detecting a reflected image of the paper sheet. The control means acquires the reflection image of the paper sheet from the image sensor, the reflection image excluding a portion where the pixel density of the transmission image exceeds a predetermined value, and the storage means The plurality of templates stored in the table are compared, and the paper sheets are identified.

  By configuring in this way, it is possible to detect the presence of a bent portion in the paper sheet from the transmission image and compare the reflected image excluding that portion with the template image, so that the discrimination accuracy can be improved. .

  In the paper sheet discrimination apparatus, the paper sheet is a bill, and the control unit determines whether or not there is a blank portion of the bill based on a reflection image detected by the image sensor, and there is no blank portion. In this case, even when the thickness of the banknote exceeds the upper limit, the inserted banknote is accepted as a normal banknote.

  By comprising in this way, even when the thickness of a banknote exceeds the upper limit, when there is no blank part, it is determined that the end is bent and can be accepted as a normal banknote.

  In the paper sheet discrimination apparatus, the control unit determines whether or not the length of the banknote is shorter than a reference value from the reflected image detected by the image sensor, and when the length is less than the reference value, Even when the thickness of the banknote exceeds the upper limit, the inserted banknote is accepted as a normal banknote.

  By comprising in this way, even when the thickness of a banknote exceeds the upper limit, when the length of the banknote is less than the reference value, the inserted banknote can be accepted as a normal banknote.

  According to the present invention, genuine paper sheets whose edges are bent can be received as normal paper sheets, and false paper sheets to which bonding has been applied are reliably rejected. That is.

  Hereinafter, preferred embodiments of the present invention will be described. FIG. 1 is a diagram illustrating a configuration of a main part of a banknote identification device (paper sheet identification device) 11 according to an embodiment. This banknote discrimination device 11 is mounted on an automatic teller machine such as a bank.

The banknote 10 thrown in from the insertion part which is not shown in figure is conveyed by the conveyance path 12. FIG.
The image sensor (image sensor) 13 includes a transmission sensor that detects a transmission image and a reflection sensor that detects a reflection image. The image sensor (image sensor) 13 outputs a reflection image and a transmission image of a banknote conveyed on the conveyance path 12 to the control unit 16.

The magnetic sensor 14 detects the magnetic data of the magnetic ink recorded on the banknote and outputs the magnetic data to the control unit 16.
The thickness sensor 15 detects the thickness of the banknote and outputs the detected thickness data to the control unit 16.

  The storage unit 17 is image data (referred to as a template) that serves as a reference for determining the denomination of the banknote, data indicating the position and magnetic amount of magnetic ink recorded for each banknote denomination, and thickness determination. The reference value etc. are stored.

  The control unit 16 determines whether or not a portion where the darkness of the pixel of the transmitted image of the banknote detected by the image sensor 13 exceeds a predetermined value exists at the end of the banknote. For example, the predetermined value serving as a criterion for determining the pixel density is determined in advance by measuring the pixel density of the transmission image when the banknotes are folded and overlapped. When there is a portion where the pixel density of the transmission image exceeds a predetermined value at the end portion, the reflected image obtained by masking the portion is compared with a plurality of templates stored in the storage unit 17 to bills Determine the denomination. Thereby, since the comparison of an image can be performed except the influence of the bending of a banknote, discrimination accuracy can be improved. In addition, it is desirable to reject the case where the area where the bill edges are folded and overlapped occupies a predetermined predetermined ratio or more of the template. This is because when there are many portions that are bent and overlapped, there are few images for comparison, and it is difficult to determine a genuine note even if the remaining portions are compared with a template.

  The control unit 16 compares the data detected by the magnetic sensor 14 with the data indicating the recording position and magnetic quantity of the magnetic ink of the corresponding denomination stored in the storage unit 17, and whether or not they match. Determine. The control unit 16 further determines whether or not the thickness of the banknote detected by the thickness sensor 15 exceeds a reference value.

  The control unit 16 discriminates the authenticity of the banknote based on the determination result of the detection data of the image sensor 13, the determination result of the detection data of the magnetic sensor 14, and the determination result of the detection data of the thickness sensor 15.

FIG. 2 is a flowchart of the discrimination process of the banknote discrimination device 11 according to the embodiment.
First, image sensor processing is executed on a bill (S11). In this image sensor process, it is determined whether or not the pixel density of the transmitted image of the banknote detected by the image sensor 13 exceeds a predetermined value. Furthermore, pattern matching between a reflected image obtained by removing a portion where the pixel density of the transmitted image exceeds a predetermined value from the reflected image detected by the image sensor 13 and a plurality of templates stored in the storage unit 17. The denomination of the template having a high coincidence rate is determined as the denomination of the inserted bill.

3A and 3B are explanatory diagrams of conventional pattern matching and pattern matching according to the embodiment.
In the conventional method, as shown in FIG. 3A, pattern matching is performed between the entire original data (reflected image) detected by the image sensor and a plurality of templates serving as a denomination determination criterion.

  As shown in FIG. 3B, the method according to the embodiment includes a reflection image excluding a portion where the pixel density of the transmission image exceeds a predetermined value, and a plurality of templates stored in the storage unit 17. The banknote denomination is determined by performing pattern matching.

  In this way, by masking the portion where the pixel density of the transmission image exceeds a predetermined value (estimated as the folded portion of the bill), the portion that does not match the template when pattern matching is performed is previously removed. Image comparison. As a result, the possibility of non-coincidence due to pattern matching due to the influence of bending of banknotes can be reduced, so that the discrimination accuracy can be further increased.

  Next, magnetic sensor processing is executed (S12). The storage unit 17 stores data indicating the position where the magnetic ink is recorded and the amount of magnetism in association with the state of banknotes for each denomination (upward, downward, front, back, etc.). In the magnetic sensor process of step S12, these data stored in the storage unit 17 are compared with the data detected by the magnetic sensor 14, and the authenticity of the banknote is discriminated based on whether or not they match.

Next, a thickness sensor process is executed (S13). In this thickness sensor process, it is determined whether the thickness of the banknote exceeds the upper limit value.
Next, it is determined whether or not the results of all sensor processes are normal (S14). When the detection data of all the sensors are normal (S14, YES), the process proceeds to step S15, and the inserted banknote is accepted as a normal banknote.

  On the other hand, the authenticity of the bill is normal, but when a thickness abnormality is detected by the thickness sensor process (S14, NO), the process proceeds to step S16, and the density of the pixel of the transmitted image at the end of the bill is predetermined. Determine whether the value is exceeded.

  When the thickness of the banknote exceeds the upper limit and the darkness of the pixel of the transmission image at the edge of the banknote exceeds a predetermined value (S16, YES), it is determined that the banknote is broken, It progresses to step S15 and is received as a normal banknote.

When the darkness of the transmitted image at the edge of the bill is equal to or less than the predetermined value (S16, NO), the process proceeds to step S17 to determine whether or not the bill has a blank portion.
If the thickness of the banknote exceeds the upper limit and there is no margin in the banknote, it is determined that there is no margin because the banknote is broken, and the process proceeds to step S15 and is accepted as a normal banknote.

When there is a blank part in the bill (S17, YES), the process proceeds to step S18, and it is determined whether or not the length of the bill in the longitudinal direction is shorter than the specified length.
When the thickness of the banknote exceeds the upper limit value and the length of the banknote in the longitudinal direction is shorter than the specified length (S18, YES), it is determined that there is a fold and the process proceeds to step S15 and is accepted as a normal banknote.

  When the thickness of the banknote exceeds the upper limit and the length of the banknote in the longitudinal direction is longer than the specified length (S18, NO), there is a possibility that the banknote is a forged banknote. Reject as an appropriate banknote.

  In the above processing, when the pixel of the transmission image at the edge of the banknote exceeds the predetermined value, when there is no margin of the banknote, or when the banknote is shorter than the specified length, it is accepted as a normal banknote. The conditions may be narrowed down, for example, only when the darkness of the pixel of the transmission image at the edge of the banknote exceeds a predetermined value, it may be accepted as a normal banknote.

  According to the above-described embodiment, even when the thickness of the banknote exceeds the upper limit value, if the density of the pixel of the transmission image at the edge of the banknote exceeds a predetermined value, the original blank portion exists in the banknote. If not, or if the bill is shorter than the specified length, it can be accepted as a normal bill. Thereby, the forged banknote which accepted the regular banknote with a fold and cut the blank paper, etc. can be rejected. By doing so, it is possible to reduce the number of folded banknotes being rejected by an automatic teller machine or the like, and this is particularly effective in overseas countries where many bent banknotes are distributed.

  Paper sheets are not limited to banknotes but may be securities certificates or the like. The device on which the bill validator is mounted is not limited to an automatic teller machine, and may be another device such as a vending machine.

It is a figure which shows the structure of the banknote discrimination apparatus of embodiment. It is a flowchart of the discrimination process of embodiment. It is explanatory drawing of the pattern matching of the conventional pattern matching and embodiment. It is a flowchart of the conventional discrimination process. It is a figure which shows the banknote with a forged banknote and a bending part.

Explanation of symbols

DESCRIPTION OF SYMBOLS 11 Banknote discrimination apparatus 12 Conveyance path 13 Image sensor 14 Magnetic sensor 15 Thickness sensor 16 Control part 17 Memory | storage part

Claims (5)

A thickness sensor for detecting the thickness of the paper sheet;
An image sensor for detecting a transmission image of the paper sheet;
When the thickness sensor detects that the thickness of the paper sheet exceeds an upper limit, and further, when the image sensor has a pixel density of a transmission image at the end of the paper sheet exceeding a predetermined value If detected, a paper sheet discrimination device comprising: a control unit that determines that the end of the paper sheet is bent and receives the input paper sheet as a normal paper sheet. Storage means for storing a plurality of templates serving as criteria for determining the type of paper sheet;
The image sensor has a function of detecting a reflected image of the paper sheet,
The control means acquires the reflection image of the paper sheet from the image sensor, and stores the reflection image excluding a portion where the pixel density of the transmission image exceeds a predetermined value, and the storage means The paper sheet discrimination device according to claim 1, wherein the paper sheets are discriminated by comparing the plurality of templates.   The paper sheet is a bill, and the control means determines whether or not there is a blank portion of the bill based on the reflected image detected by the image sensor, and if there is no blank portion, the thickness of the bill The paper sheet discriminating device according to claim 1 or 2, wherein the inserted bill is accepted as a normal bill even when the amount exceeds the upper limit.   The control means determines whether or not the length of the banknote is shorter than a reference value from the reflected image detected by the image sensor, and when the length is less than the reference value, the thickness of the banknote has an upper limit value. The paper sheet discrimination device according to claim 3, wherein the inserted bill is accepted as a normal bill even when the bill is exceeded. A paper sheet discrimination method of a discrimination device for discriminating paper sheets,
The thickness sensor detects the thickness of the paper sheet,
Detecting a transmission image of the paper sheet with an image sensor,
The thickness sensor detects that the thickness of the paper sheet exceeds an upper limit value, and the image sensor has a pixel density of a transmission image at an end of the paper sheet exceeding a predetermined value. Is detected, it is determined that there is a fold at the end of the paper sheet, and the inserted paper sheet is accepted as a normal paper sheet.

Images