JP3651177B2 - Paper sheet identification device - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a paper sheet discrimination apparatus that discriminates the authenticity of a paper sheet, and more particularly to a paper sheet discrimination apparatus having a function of measuring the timing of detection of a conveyed paper sheet.
[0002]
[Prior art]
In recent years, forgery of paper sheets is often performed by cutting a copy of the paper sheets. The fake forged paper sheets made in this way are difficult to detect by partial linear scanning with a point sensor, and a surface pattern sensor that can read the pattern of the surface of the paper sheet is used. It is effective for distinguishing cut and forged paper sheets. Therefore, a surface pattern sensor is used for the highly accurate and highly functional paper sheet discriminating section to read the entire pattern of the target paper sheet.
[0003]
By the way, as this surface pattern sensor, a contact type linear image sensor type or a lens reduction type linear image sensor type may be used.
[0004]
6A and 6B are conceptual diagrams illustrating a conventional configuration example of a lens pattern reduction type linear image sensor type surface pattern sensor, where FIG. 6A is a diagram viewed from a direction perpendicular to the paper sheet conveyance direction, and FIG. It is the figure seen from the direction. The surface pattern sensor 60 mainly includes a linear image sensor 61 arranged in a direction perpendicular to the conveyance direction of the paper sheet 64, a light source 62 that illuminates the paper sheet 64, and linearly transmitted light from the paper sheet 64. An imaging lens 63 that forms an image on the image sensor 61 is formed.
[0005]
A position detection sensor 65 is provided immediately before the surface pattern sensor 60 in order to measure the timing at which the conveyed paper sheet 64 enters the scanning line of the surface pattern sensor 60. The position detection sensor 65 detects the presence or absence of the target paper sheet 64. The surface pattern sensor 60 starts scanning of the linear image sensor 61 in response to a signal from the position detection sensor 65. In this way, the pattern of the surface of the paper sheet 64 is identified.
[0006]
On the other hand, there is a method in which the position of the paper sheet is detected without using the position detection sensor 65 and the pattern on the surface of the paper sheet 64 is identified.
In this method, when a paper sheet is inserted into the input port of the discriminator, it is detected that the paper sheet has been inserted by a paper sheet insertion detection sensor in the vicinity of the input port. The transport mechanism is driven and the paper sheets begin to be transported. The scanning of the surface pattern sensor is started before the paper sheet enters the scanning line of the surface pattern sensor in any case.
[0007]
However, in this case, since the surface pattern sensor starts scanning before the paper sheet enters the scanning line, the image data of the background and the paper sheet is captured without distinction. Therefore, this image data is temporarily stored in the memory, the outer edge portion of the paper sheet is extracted by special image processing, and the paper sheet portion or a part thereof is extracted from the captured image data. It is used as data for authenticity discrimination.
[0008]
As an example using such a discrimination means, there is Japanese Patent Application No. 8-174465 already filed by the applicant of the present application. Here, for each scan of the surface pattern sensor, the difference in output between adjacent pixels is taken and the absolute value or square value is obtained. That is, image data for one scan is differentiated in the pixel arrangement direction, and an absolute value or a square value of the value is obtained. Hereinafter, the value obtained by differentiation and squaring is referred to as “differential square value”.
[0009]
FIG. 7 is a diagram showing the relationship between the output of the surface pattern sensor and the differential square value. In the figure, the characteristic M10 is the output of the surface pattern sensor, and here, as a paper sheet, for example, a black line and a blue line written on a blank sheet surface are used. As indicated by the characteristic M10, the output value hardly changes in the pixel detecting the blank portion. The output of the pixel detecting the black line is greatly reduced as in the portion M10a. Further, the pixel detecting the blue line is reduced by an intermediate size between the portion M10a and the blank sheet detection unit as in the portion M10b.
[0010]
Further, a lower characteristic L10 in FIG. 7 is a graph obtained by calculating a differential square value of the characteristic M. In this case, the portion where the blank paper is detected has a differential square value of almost “0”. On the other hand, the differential square values of the part M10a and the part M10b have large differences in values as indicated by the characteristic L10a and the characteristic L10b, respectively. Thus, by obtaining the differential square value, the change in the pattern can be clearly discriminated.
[0011]
However, if the linear image sensor has a high resolution only by performing pattern matching based on this differential square value, there is a possibility that an erroneous determination will be made due to a subtle displacement of the object. Therefore, in Japanese Patent Application No. 8-174465, image data is divided into regions having arbitrary sizes, and the sum of differential square values included in each region is obtained. Hereinafter, this total value is referred to as a differential square sum. By calculating the differential sum of squares, it is possible to quantitatively evaluate the contrast of the image in the region, to reduce erroneous discrimination due to the displacement of the object, and to reduce the contrast as in a black and white copy. When the difference is high, the differential square sum is a large value. On the other hand, when the contrast is low, such as blank paper, the differential square sum is a small value, and it is possible to discriminate from a real paper sheet.
[0012]
[Problems to be solved by the invention]
However, the conventional method has the following drawbacks.
(1) First, as shown in FIG. 6, in the case where one or several position detection sensors are used to detect paper sheets entering the scanning line of the surface pattern sensor, the number of parts is increased by including a detection circuit or the like. There is a problem that the cost increases.
(2) On the other hand, in Japanese Patent Application No. Hei 8-174465, scanning is started before paper sheets enter the scanning line of the surface pattern sensor, so image data including the background is temporarily stored in the memory. In addition, it is necessary to extract the outer edge portion of the paper sheet by image processing or the like in the subsequent stage. For this reason, there is a problem that the amount of image data to be handled increases, the load on the outer edge portion extraction processing also increases, the processing time becomes longer, or a high-speed and high-performance processor is required, resulting in an increase in cost.
[0013]
The present invention has been made in view of the above points, and an object thereof is to provide a paper sheet discrimination apparatus that is low in cost and can shorten the discrimination time.
[0014]
[Means for Solving the Problems]
In the present invention, in order to solve the above-mentioned problem, in a paper sheet discrimination apparatus for identifying the authenticity of a paper sheet, a linear image sensor arranged in a direction perpendicular to the transport direction of the paper sheet, and the paper sheet A surface pattern sensor having a light source for irradiating transmitted light, an imaging lens for forming an image of the transmitted light from the paper sheet on a linear image sensor, and the linear image sensor as the paper sheet is conveyed. A pattern detection unit that scans at a predetermined time interval to detect a pattern on the surface of the paper sheet, and obtains an output difference between pixels of the linear image sensor every predetermined scan, and an absolute value of each output difference or And an entry detecting means for detecting that the leading portion of the conveyed sheet has entered the scanning line of the surface pattern sensor based on a sum of square values. A classifier is provided
[0015]
In such a paper sheet discrimination apparatus, when the pattern detection means scans the linear image sensor at regular time intervals as the paper sheet is conveyed and detects the pattern of the surface of the paper sheet, the entry detection means The output difference between the pixels of the linear image sensor is obtained every predetermined scan, and the leading portion of the conveyed paper sheet is scanned by the surface pattern sensor based on the absolute value of each output difference or the sum of the square values. Detect that you entered the line.
[0016]
That is, since the light source is directly viewed before the paper sheet enters the scanning line, the value of the output difference between the pixels of the linear image sensor for each scan increases, and the absolute value or square value of the output difference. The total value of becomes larger. On the other hand, in the scanning immediately after the paper sheet enters the scanning line, the outer edge portion of the paper sheet is read. At this time, there is no normal pattern on the outer edge portion, and the light is scattered and uniformly incident on the linear image sensor, so that the value of the output difference between the pixels becomes small, and the sum of the absolute value or square value of the output difference. The value becomes smaller.
[0017]
By detecting the difference in output difference in this way, it becomes possible to detect that the outer edge of the paper sheet has entered the scanning line of the surface pattern sensor without using a position detection sensor as in the prior art. Therefore, the structure is simplified.
[0018]
Further, it is not necessary to perform complicated processing such as temporarily storing image data including the background in the memory and extracting the outer edge portion of the paper sheet by image processing or the like.
[0019]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, an embodiment of the present invention will be described with reference to the drawings.
FIG. 1 is a conceptual diagram of the function of the paper sheet discrimination apparatus of this embodiment. A light source 2 is provided below the conveyance surface of the paper sheet 1. The light source 2 irradiates the paper sheet 1 with the output light 2a. On the other hand, on the upper side of the conveyance surface, a light receiving unit comprising a linear image sensor arranged in a direction perpendicular to the conveyance direction and an imaging lens that forms an image on the linear image sensor of the output light 2a transmitted through the paper sheet 1. 3 is provided. The light source 2 and the light receiving unit 3 constitute a surface pattern sensor 4.
[0020]
The pattern detecting means 5 intermittently detects the pattern of the surface of the paper sheet 1 by scanning the linear image sensor in the light receiving unit 3 at regular time intervals as the paper sheet 1 is conveyed. And the approach detection means 6 calculates | requires the output difference between each pixel in the predetermined range of a linear image sensor for every predetermined scanning, and calculates | requires the absolute value of each output difference, or the total value of a square value. Based on this total value, it is detected that the leading portion 1 a of the conveyed paper sheet 1 has entered the scanning line of the surface pattern sensor 4.
[0021]
FIG. 2 is a diagram showing a schematic configuration inside the paper sheet discrimination apparatus according to the present embodiment. Here, the example which discriminates a banknote, for example is shown as paper sheets. Below the conveyance surface of the banknote 11, the LED array 12 as a light source is provided. On the other hand, above the conveyance surface, a linear image sensor (not shown) arranged in a direction perpendicular to the conveyance direction, and an imaging lens (not shown) that forms an image of the light 12a that has passed through the banknotes on the linear image sensor, A light receiving unit 13 is provided. The LED array 12 and the light receiving unit 13 constitute a surface pattern sensor 14.
[0022]
The control device 15 is composed of a processor, a memory, and the like, and discriminates the banknote 11 according to a built-in program. When a detection signal is sent from the insertion detection sensor 16 provided at the bill insertion slot of the paper sheet discrimination device, the control device 15 operates a transport mechanism unit (not shown). At the time when the banknote 11 is inserted, the banknote 11 has not yet reached the position of the surface pattern sensor 14, but the control device 15 starts sensing by the surface pattern sensor 14. However, the image data is not written in the memory until the leading portion 11a of the banknote 11 is detected by a method described later.
[0023]
The control device 15 detects the output of each pixel of the built-in linear image sensor from the surface pattern sensor 14 simultaneously with the start of sensing. And the control apparatus 15 calculates the square value of the output difference between adjacent pixels from the output value for every scanning. That is, the image data for one scan is differentiated in the pixel arrangement direction, and the square value of the value is obtained. Hereinafter, the value obtained by differentiation and squaring is referred to as “differential square value”.
[0024]
Further, the control device 15 obtains the total value of the differential square values for the pixels for one scan or the limited portion of the pixels. Hereinafter, this total value is referred to as differential square sum. Then, the leading portion 11a of the banknote 11 is detected based on this differential square sum.
[0025]
Next, a specific method for detecting the leading outer edge portion 11a of the banknote 11 will be described.
First, before the banknote 11 enters on the scanning line of the surface pattern sensor 14, the linear image sensor in the light receiving unit 13 directly views the output light of the LED array 12. FIG. 3 is a diagram illustrating output characteristics of the linear image sensor when the output light of the LED array 12 is directly viewed. Here, the characteristic M1 indicates an actual output from the linear image sensor. A characteristic L1 indicates a differential square value of an output from the linear image sensor. Here, although 50 pixel data are shown, in reality, the linear image sensor is provided with more, for example, 178 pixels.
[0026]
In the characteristic M1, the portions M1a, M1b, and M1c protruding downward (brighter) are the portions where the output light of the LED array 12 is directly viewed. The differential square values of the portions M1a, M1b, and M1c are large values unlike the other portions as seen in the corresponding portions L1a, L1b, and L1c of the characteristic L1.
[0027]
In this way, the control device 15 sums up the differential square value obtained by the characteristic L1 for pixels of one scan or a limited portion, and obtains the differential square sum. Here, a differential square sum is obtained for the illustrated pixels (0 to 50), and the differential square sum at this time is 27112. Note that the value of the differential square sum shown here is a value obtained by calculating the output of the pixel based on digital data (for example, 8-bit data).
[0028]
On the other hand, FIG. 4 is a diagram showing output characteristics of the linear image sensor when the banknote 11 enters the scanning line of the surface pattern sensor 14. The leading portion 11a of the banknote 11 is usually a margin. That is, as shown in FIG. 5, since the banknote 11 has the margin 11c at the periphery of the pattern surface 11b, the output light of the LED array 12 is scattered by the margin 11c and uniformly incident on the light receiving unit 13. Therefore, when the banknote 11 enters the scanning line of the surface pattern sensor 14, the output of the linear image sensor has a small output difference between pixels as seen in the characteristic M2 in FIG. As a result, as can be seen from the characteristic L2, the differential square value is not “0” due to variations in the sensitivity of the sensor, non-uniformity of the light source, and the like, but when the light source is viewed directly or the pattern surface 11b is scanned. The value is very close to “0” compared to the case of The differential square sum in this case is 63, for example.
[0029]
Each time the control device 15 obtains the differential square sum, it compares the value with a preset threshold value. If the differential square sum is smaller than the threshold value (for example, 1000), the margin 11c is set. Judge that it is detected. That is, the control device 15 determines that the leading portion 11 a of the banknote 11 has entered the scanning line of the surface pattern sensor 14 when the differential square sum is changed from 27112 to 63.
[0030]
When the leading portion 11a of the banknote 11 is thus detected, the control device 15 thereafter starts to store image data in the memory. Thus, the detection of the pattern surface 11b is executed. In addition, the discrimination method of the banknote 11 by the detection of the pattern surface 11b may be as usual.
[0031]
As described above, in this embodiment, the differential square value between adjacent pixels is obtained from the output value for each scan, the differential square sum is further obtained, and the bill 11 is calculated based on the value of the differential square sum. Since the leading portion 11a is detected, a position sensor for detecting the position of the banknote 11 is not required, and image data is not stored in the memory before the banknote 11 enters the scanning line of the surface pattern sensor 14. The leading portion 11a of the banknote 11 can be detected. Therefore, the number of parts can be reduced and the cost can be reduced. In addition, since the image data to be handled can be reduced, the memory capacity can be reduced, the cost can be reduced, the load on data processing can be reduced, and the processing time can be shortened.
[0032]
In this embodiment, the sum of squared derivatives is obtained, but it may be a sum of absolute values of the differentiated values.
In this embodiment, the output light 12a of the LED array 12 is transmitted through the banknote 11 and received by the light receiving unit 13. However, the LED array 12 is provided on the same side as the light receiving unit 13, and the output light 12a is supplied to the banknote. 11, and the reflected light may be received by the light receiving unit 13. However, in this case, the output light of the LED array 12 is provided on the back surface of the banknote 11 by providing a reflecting plate having an appropriate pattern with high contrast (for example, a bar code having a resolution of the surface pattern sensor 14) or a mirror. It is necessary to provide something that directly reflects 12a.
[0033]
Moreover, although the example which discriminate | determines the banknote 11 as a paper sheet was shown in this form, it can also discriminate | determine with securities.
Furthermore, in this embodiment, the differential square sum is obtained for each scan, but the present invention is not limited to this, and it may be obtained for every two or more scans.
[0034]
【The invention's effect】
As described above, in the present invention, the linear image sensor is scanned at regular time intervals as the paper sheet is conveyed to detect the pattern of the surface of the paper sheet, and the pixels of the linear image sensor are detected every predetermined scan. Since the output difference is obtained, and based on the total value of the absolute value or square value of each output difference, it is detected that the leading portion of the conveyed paper sheet has entered the scanning line of the surface pattern sensor. It becomes possible to detect that the outer edge of the paper has entered the scanning line of the surface pattern sensor without using the position detection sensor. Therefore, the structure is simplified and the cost is reduced.
[0035]
Further, it is not necessary to perform complicated processing such as temporarily storing image data including the background in the memory and taking out the outer edge portion of the paper sheet by image processing or the like. For this reason, discrimination time is shortened.
[Brief description of the drawings]
FIG. 1 is a conceptual diagram of functions of a paper sheet discrimination device according to the present embodiment.
FIG. 2 is a diagram showing a schematic configuration inside a paper sheet discrimination apparatus according to the present embodiment.
FIG. 3 is a diagram showing output characteristics of the linear image sensor when the output light of the LED array is directly viewed.
FIG. 4 is a diagram illustrating output characteristics of a linear image sensor when a bill enters a scanning line of a surface pattern sensor.
FIG. 5 is a diagram showing a schematic configuration of a bill surface.
6A and 6B are conceptual diagrams illustrating a conventional configuration example of a lens pattern reduction type linear image sensor type surface pattern sensor, in which FIG. 6A is a diagram viewed from a direction perpendicular to the sheet conveyance direction, and FIG. It is the figure seen from the conveyance direction.
FIG. 7 is a diagram showing a relationship between an output of a surface pattern sensor and a differential square value.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 1 Sheet | seat 1a Lead part 2 Light source 2a Output light 3 Light receiving part 4 Surface pattern sensor 5 Pattern detection means 6 Incoming detection means 11 Bill 11a Leading part 12 LED array 12a Output light 13 Light receiving part 14 Surface pattern sensor 15 Control device

Claims (4)

In a paper sheet discrimination device that discriminates the authenticity of paper sheets,
A linear image sensor arranged in a direction perpendicular to the transport direction of the paper sheet, a light source for irradiating the paper sheet with transmitted light, and a result of imaging the transmitted light from the paper sheet on the linear image sensor. A surface pattern sensor having an image lens;
A pattern detection means for detecting the pattern of the surface of the paper sheet by scanning the linear image sensor at regular time intervals as the paper sheet is conveyed,
The output difference between the pixels of the linear image sensor is determined for each predetermined scan, and the leading portion of the conveyed paper sheet is based on the absolute value or the sum of the square values of the output differences. An entry detection means for detecting entry into the scanning line of the surface pattern sensor;
A paper sheet discrimination device characterized by comprising: 2. The paper sheet discrimination apparatus according to claim 1, wherein the entry detection unit is configured to obtain the output difference for each of the scans. The paper sheet discrimination apparatus according to claim 1, wherein the paper sheet is a banknote. In a paper sheet discrimination device that discriminates the authenticity of paper sheets,
A linear image sensor arranged in a direction perpendicular to the conveying direction of the paper sheet, a light source for irradiating the paper sheet with light, and an image forming an image of reflected light from the paper sheet on the linear image sensor A surface pattern sensor having a lens;
A reflector provided on the back side of the paper sheet as viewed from the light source, having a suitable pattern with high contrast, or having a surface that directly reflects light from the light source;
A pattern detection means for detecting the pattern of the surface of the paper sheet by scanning the linear image sensor at regular time intervals as the paper sheet is conveyed,
The output difference between the pixels of the linear image sensor is determined for each predetermined scan, and the leading portion of the conveyed paper sheet is based on the absolute value or the sum of the square values of the output differences. An entry detection means for detecting entry into the scanning line of the surface pattern sensor;
A paper sheet discrimination device characterized by comprising:

Images