JPH01276388A - Reference point determining system and papers distinguishing system using it - Google Patents

Abstract

PURPOSE:To always compare with the true ticket data of a reference position and to minimize a storage capacity to store the true ticket data by detecting one point on the side of papers among measuring data and calculating the inclination of the papers with this point as a reference. CONSTITUTION:The outline of papers extracted from measuring data is differed two times and the inflection point of the outline is detected. The coincident point of the segment to link a detected inflection point and the outline of the papers is detected, the inflection point linked by the segment, in which the coincident point is the most, is made into one point on the side of the papers and one point on the side is made into a reference point to calculate the inclination from the reference position of the papers. Thus, by comparing with the true ticket data at the reference position, the papers can be distinguished and the capacity of the true ticket data can be minimized.

Description

[Detailed description of the invention] [Table of contents] Overview Industrial application field Means for solving the problem to be solved by the invention (i) First invention (11) Second invention action (1) First invention (11) Second invention embodiment 1, correspondence relationship between the embodiment and FIG. (iii) Configuration of storage unit (iv) Configuration of determination unit (v) Configuration of central processing unit ■, Operation of the example (1) Operation of determining the reference point (ii) Overall operation ■, Summary of the example Vl Effects of the invention in terms of variations of the invention (Summary) Regarding a method for centralizing a reference point in paper sheet discrimination and a paper sheet discrimination system using the same, it is unnecessary to store a large amount of genuine bill data. With the purpose of Detect the point on the side of the paper sheet based on the number of detected matching points, and use the point on the side as the reference point for calculating the inclination of the paper sheet. do.

In addition, using this method, a measuring means for determining the desired physical quantity in each minute area of paper sheets, a storage means for storing the measured values, A processing means for performing coordinate transformation, and a genuine note data recording means for recording the measured values of the genuine note at the reference position 2 Comparison means for comparing the transformation data coordinate-transformed by the processing means and the genuine note data. and a discrimination section for discriminating paper sheets.

[Industrial application field]

The present invention relates to a method for determining a reference point when identifying paper sheets such as securities, and a paper sheet identification system using the method.

[Conventional technology]

In a system for differentiating paper sheets such as banknotes and securities, the entire paper sheet to be inspected (for example, securities) is divided into minute areas, and each minute area is inspected using an optical sensor, a thickness sensor, etc. It measures the intensity, color, and external shape of the pattern, converts the measured values into gradation values, etc., and stores them. The securities are identified based on the results of comparing this measured value with genuine bill data that has been measured and stored in advance.

5M is a configuration diagram of a conventional securities identification system 1.

Hereinafter, the securities subject to identification will be simply referred to as securities.

The conveyance unit 503 of the measurement unit 500 carries the image sensor 501
The security is conveyed to the sensing range of the image sensor 501 so that the gradation of the entire security can be measured.

The image sensor 501 senses the gray scale of each minute area on the surface of the security to which each element corresponds, and converts it into an electrical signal. The electrical signal is analog-digital (A
/I)) The converted value is converted into a digital value representing the gradation by the converter 502 and stored in the measurement data storage area 521 of the storage unit 520.

The coordinates of each minute area of the security are expressed with the two-dimensional elements of the iron and image sensor 501 arranged in the X direction and Y direction, respectively, and the measurement data is an element of the array corresponding to the coordinates of the minute area. shall be stored in .

The transport unit 503 generally transports the securities so that the long sides are generally in the Y direction, but due to misalignment that occurs during transport, the long sides of the securities are tilted with respect to the Y axis and the securities are transported from innocent to forward. 50
It is also conceivable that measurement using 1 is performed. The measurement data measured in such a tilted state will be referred to as skew data hereinafter.

The genuine bill data storage area 531 stores the results of measuring a genuine bill using the same image sensor 501 (genuine bill data) in order to compare the measured data to identify the securities. Since the inclinations of the skew data vary, in order to compare this, genuine bill data at all inclination angles are required.

The central processing unit (CPU) 510 determines the inclination angle θ of the security from the measurement data, searches the genuine note data storage area 531 for genuine note data measured at the inclination angle θ, and uses this as genuine note data for comparison. The discriminating unit 530 is instructed to use it as a.

The discriminating unit 530 compares the measured data of the securities stored in the measured data storage area 521 with specified genuine bill data, identifies the securities based on the result, and outputs the data from the input/output unit 540 such as a display or printer. Output and report the identification results.

[Problem to be solved by the invention] By the way, in the conventional securities identification system mentioned above,
In order to directly compare the skew data with genuine bill data, it is necessary to store genuine bill data measured at all tilt angles in the genuine bill data storage area 531. Therefore, there was a problem in that a huge amount of capacity was required to store genuine bill data.

The present invention was created in view of these points, and is a paper sheet @ designed to reduce the capacity of genuine bill data.
The purpose is to provide an identification system.

[Means for Solving the Problem] (i) X1 Critical Point FIG. 1(A) is a block diagram of the principle of the reference point determination method of the present invention.

First, the outline of the paper sheet is extracted.

An inflection point is detected by performing subtraction twice on the extracted contour.

A matching point between the line segment connecting the detected plurality of inflection points and the contour is detected.

Based on the number of detected matching points, points on the edges of the paper sheet are identified.

In this way, the reference point for calculating the inclination of the paper sheet is determined.

(11) Second Injection FIG. 1(B) is a block diagram of the principle of the paper sheet discrimination system of the present invention.

In the figure, a measuring means 104 measures a desired physical quantity in each minute area of the paper sheet.

The storage means 105 stores the measured values of physical quantities of paper sheets measured by the measuring means 104.

The processing means 106 determines a reference point for calculating the inclination of the paper sheet from the measured value stored in the storage means 105, calculates the necessary correction amount of the inclination, and uses the measured value stored in the storage means 105. Rearrange and perform coordinate transformation.

The discriminating unit 107 includes a genuine bill data recording means 171. [Genuine bill data recording means] Comparison means 172 is provided for comparing the genuine bill data recorded in the genuine bill data 71 and the transformation data obtained by coordinate transformation by the processing means 106, and the paper sheets are discriminated.

Therefore, overall, the 4Qp point is determined, the amount of correction for the inclination of the paper sheet is calculated, the measured value is coordinate-transformed to the reference position,
It is configured to compare paper sheets with genuine bill data located at the reference position Lj to identify paper sheets.

[Work] ”” A clever invention Hereinafter, reference will be made to FIG. 1 (Δ).

The inflection point of the outline is detected by subtracting twice the circle 7 of the paper sheet extracted from the measurement data.

Detect a matching point between the line segment connecting the detected inflection points and the outline of the paper sheet, and set the inflection point connected by the line segment that has the most matching points as one point on the edge of the paper sheet, One point on this side is used as a reference point for calculating the inclination of the paper sheet from the reference position.

"Escape 1)■ Hereinafter, reference will be made to FIG. 1(B).

The measurement values in each minute region of the paper sheet measured by the measuring means 104 are stored in the storage means 105.

A reference point is determined from the measurement data stored in the storage means 105 according to the reference point determination method, and the inclination of the sheet from the reference position is calculated. The processing means 106 performs coordinate transformation to correct the inclination of the paper sheet, and converts the measurement data into transformation data corresponding to the case where the paper sheet is measured at a reference position.

The sorting stage 172 of the discriminating unit 107 compares the genuine bill data with the converted data after the coordinate transformation to discriminate the paper sheets.

In the present invention, since the measured data is corrected and compared with genuine note data, there is no need to prepare genuine note data in accordance with a large number of skew data.

[Example] Hereinafter, an example of the present invention will be described in detail based on the drawings.

The second screen 31 shows the configuration of a +jlf bill validation system according to an embodiment of the present invention.

-7-α Teruenji Masutsuzume V (Reception-Appendix-Related Matters) Here, the correspondence between the embodiments of the present invention and FIG. 1 will be shown.

Measuring means 104 LJ corresponds to the measuring section 200.

The storage means 105 corresponds to the storage section 220.

The processing means 10G is a central processing unit (CI) U) 210
corresponds to

The determining unit 107 corresponds to the determining unit 230.

Genuine bill data recording means 171 stores genuine bill data storage area 23
] corresponds to

The comparison means 172 +1 corresponds to the comparison section 232.

Embodiments of the present invention will be described below assuming that there is a 4V'' correspondence as described above.

↓-Tinitis p-trap FIG. 6 is an explanatory diagram of the principle of coordinate transformation.

In Figure (a), the rectangle representing the security is tilted by θ with respect to the xY coordinate axis, and is in an oblique state.

At this time, the coordinate axes and origin can be rearranged to conveniently represent the coordinates of each point in the rectangle, and the coordinates of each point in the long sword shape can be transformed.

For example, by performing coordinate transformation such that the origin is at the vertex and the coordinate axes are at side A B and side AC,
Converting the 6-position i6 relationship in which the measurement data is tilted at an angle θ i with respect to the XY coordinate axes as shown in a) to a reference position that is not tilted with respect to the xY coordinate axes as shown in Figure 6(b). I can do it.

When measuring a security by dividing it into minute areas, the positional relationship of the elements of the array that stores the measured values in each minute area of the security,
The measurement value is configured to be stored in the storage area so that the positional relationship of each minute area corresponds. In this way, the elements of the storage area array that correspond to the measurement positions are called pixels,
Coordinate transformation can be performed by rearranging these pixels, such as converting the measured data of the skewed security into converted data equivalent to the measured data when measured at the reference position.

So, in order to perform the coordinate transformation as described above, 1,
It is necessary to calculate the slope θ without showing any deviation from the reference position of the security. This inclination θ is determined by specifying a pixel on one of the four sides of the ring 7I (of the security) as a reference point, and using this as a reference to determine the continuous state of the coordinates of the pixels that make up the contour of the security. It can be calculated based on each investigation.

In Figure 2, the securities identification system
A measurement unit 200 that measures the density gradation in the 1■ area, a storage unit 220 that stores measurement data, etc., and a central processing unit ((,, PU) that processes the measurement data and converts it into conversion data.
210, and a determination unit 2 that determines whether the securities are genuine or not.
30, and an input/output device 240 that outputs the inspection results of securities by the discriminating section 230.

Here, too, the securities that have undergone identification are simply referred to as securities.

The measurement unit 200 includes a two-dimensional image sensor 201 that measures the density gradation of the securities. A digital (A) converts the output of the image sensor 201 into digital data.
/D) converter 2o2. It is comprised of a conveyance unit 203 that conveys the securities within the sensing range of the image sensor 201.

Here, the two-dimensional image sensor 201 is composed of mXn elements, and the directions in which the elements are arranged are defined as an X direction and a Y direction, respectively.

The coordinates of a minute area in which a security is located within the sensing range of the image sensor sensor 201 correspond to the order in which the elements of the image sensor 201 that sense the concentration of the minute area are arranged in the X and Y directions, and the minute areas are arranged in the X and Y directions. The positional relationship between the two is saved.

Further, the Denkle value indicating the sensing output of the image sensor 201 converted by the A/D converter 202 is set to "O" when there is no sensor within the sensing range of the element of the image sensor 201, that is, when it is "white". It is expressed as a numerical value that increases as the gradation of light and shade approaches "black."

] Excretion-) Note 1: The +1 father storage unit 220 of part a is a measurement data storage area 221 that stores measured values of securities. Conversion data storage area 222 for storing conversion data that is the result of coordinate conversion of measurement data. A one-dimensional array M having m and r) elements each consisting of the coordinates of the pixels of the outline of the security. X.

Shaft push data storage 9ii1 area 223 for storing Moy.
MoX.

l′1. One-time differential data storage area 224, IVI, X, which stores one-dimensional arrays M,
It is formed as a two-time difference data storage area 225 which has m and n elements each consisting of the difference values of M and Y, and stores -911 dimensional arrays M2X and M2Y.

The measurement data is stored in the image sensor 201 such that the positional relationship between the storage locations of the outputs of each element of the image sensor '201 corresponds to the positional relationship of the elements of the image sensor '201.
The data are stored in the measurement data storage area 221 in the form of an array having ITI X n elements, which is the same as the arrangement of elements. Therefore,
The positional relationship between minute areas is determined by the measurement data storage area 221.
It is also preserved in the positional relationship of the elements of the array.

-(-iv) The Kusunoki discrimination unit 230 of the separate section III stores genuine note data in advance, which stores the results measured by the image sensor 201 with the genuine note at the reference position (hereinafter referred to as genuine note data). Territory JA23]
and a comparison unit 232 that compares the converted data and the Nanshin note data.
It is formed by combining and.

Here, the genuine note data is assumed to be measured at a reference position where the long side of the genuine note is parallel to the Y direction, for example.

(v) The central processing unit 210 includes a one-time difference processing unit 211 that differs the contents of the contour data storage area 223 and stores it in the difference data storage area 224 once, and a one-time difference data storage area 224. The contents of the meeting 11 area 224 are differentiated and stored twice in the differential data storage area 225.
J, L:
It is configured to determine the tit point, calculate the number of securities, and perform coordinate transformation of measurement data.

The one-time difference processing unit 211 uses M stored in the contour data storage area 223. Read y1M6y and write its i-
The value obtained by subtracting the first element from the element of [No. 1 1] is calculated once as 1 of M I
-1 No. 1" element.

Taking the difference between adjacent elements one after another in this way is called a difference.

2-time difference processing unit 2] 2, 1-time difference data storage area 22
4, and the results are stored twice in the differential data storage area 225.
, , is stored in .

Figure 3 is a flowchart showing the operation for determining the W single point in the embodiment shown in Figure 2.

FIG. 4 is an explanatory diagram of measurement data of securities.

Hereinafter, reference will be made to FIGS. 2 to 4.

In FIG. 4, each of the eyes of the base corresponds to an element (pixel) of an array that stores the output of one element of the image sensor 201, and the entire eye of the base corresponds to the image sensor 201.
1 to an array of m rows and n columns, which covers the entire sensing range. Further, the directions of thick arrows in the figure correspond to the X and Y directions of the arrangement of the elements of the image sensor, which correspond to the rows and columns of the arrangement, respectively.

The CPU 210 selects the first 0 of the elements in each row of the measurement data array stored in the measurement data storage area 221.
An array element having a value other than " is detected, and the column number to which the element belongs is stored in a one-dimensional array Mox having m elements in the contour data storage area 223. For example, as shown in FIG.
The area indicated by diagonal lines in Δ) is detected as a gradation change point in the χ direction. These are points on the contour of the security, and the detected elements are Mox(n, n, na.

n-a-b, n-a-2b, ... n-,
n- becomes -c, n- becomes -2c, n, n, n). Here, the number of the element of MoX, that is, the row number of the array, corresponds to the Y coordinate of the outline of the security, and the contents of the element of MoX correspond to the X coordinate (step 301).

Contour data M stored in the contour data storage area 223
. The difference processing unit 211 performs a difference on X once and stores it in a one-dimensional array MIX having m elements in the difference data storage area 224. MIX is, for example, M+x(0,0
, a, b, b, ・+, b.

c, c, +2c, 0, 0, 0) (step 30
2).

The content of MIX is further differentiated by the two-time difference processing unit 212, and the result is stored in the two-time difference data storage area 225 as a one-dimensional array M2X having m elements. For example, if M and X are further differentiated, M2X(Q, 0.-23.
r>-b, O, O2, --, b-c, 0. -to-c, k
12c, 0.0.0) (step 303).

The contents of the two-time differential data storage area 225 correspond to the two-time differential of the contour data MOX stored in the data storage area 223. Therefore, among the elements of M2X, "0"
Pixels corresponding to elements other than ``'' are inflection points of the outline of the security.For example, pixels P and Q in FIG.

Four points R and S are detected as inflection points. (Step 3
04).

The CPU 210 obtains an equation for a line segment connecting two of the inflection points detected in step 304. For example, in Figure 4 (
If the four points P, Q, R, and S in a) are detected as inflection points, the line segment connecting P and Q points, the line segment connecting P and R points, and the line segment P and 8 Line segment connecting points ■, Line segment connecting point Q and R point ■, Line segment connecting point Q and point 8 ■ Line segment connecting point 2R and point 8 ■
d+ X+e+ Y=f+ ” '■dz X−
) e 2 y= r2 ” ' ■d:
+ X+e:+ Y=f:+ ---■da Xl-
04Y=fs' ・-■d-,X+-es Y=f
s...■(1,X-1-e, Y-(b
・ ・ ・Determine as shown in ■ (step 305).

The CPU 210 receives contour data M. The element of X is the X coordinate,
M substituting the number of the element of MOX as the Y coordinate into the equation obtained in step 305 to make each equation valid.

Count the number of elements in X. Here, M is the number that satisfies the i-th equation. The number of elements of X is B.

Let 3 be a one-dimensional array with B as an element.

For example, the equation representing the line segment ■ is M, which corresponds to the diagonally shaded pixel between points P and Q in FIG. It holds true for p elements of

Therefore, B, -p, B2-2 (step 306)
.

In step 306, the largest number of M. Let the end points of the line segment for which the equation holds true for the elements of X be points on the edge of the security. For example, if a one-dimensional array B is B (p, 2, 2.
q, 2. r), if p < r < q, then point Q and point R, which are the points at both ends of the line segment ■ corresponding to q, are points on the side of the security, so they are diagonal. Use as a reference point to calculate the slope of the security (step 307)
.

In the same way, the process of step 301 is performed for each column of the array, and the change point of the gradation in the Y direction shown with diagonal lines in FIG. 1 with n elements in the contour data storage area 223.
Dimensional array M. Store in 7.

Next, as in steps 302, 303, and 304, the inflection points P, Q, and T are detected, and as in step 305, a line segment connecting point P and point T, and point Q and point Find the equation of the line segment ■ connecting T, and add M to each equation according to step 306. Assign the number of the element 7 as the X coordinate and the content of the element May as the Y coordinate, and select the largest number M. Let the inflection point connected by the equation that holds true for the elements of , be one point on the side of the security.

(ii) The density gradation of each minute region of the security measured by the overall motion measurement unit 200 is stored in the measurement data storage area 221 of the storage unit 220.

The CPU 210 stores the measurement data in the measurement data storage area 22.
1, determine the reference point for calculating the inclination as described above, and calculate the inclination θ, which is the correction amount for correcting the measurement data to the reference position.

The CPU 210 performs coordinate transformation so that each vertex of the security in the measurement data becomes a reference position, corrects the inclination θ, and stores the resulting transformation data in the transformation data storage area 222.

The determining unit 230 compares the genuine note data read from the genuine note data storage area 231 with the converted data read from the converted data storage area 222 using the comparing unit 232, and determines whether the security is a genuine note based on the result. The input/output device 240 reports the result of the discrimination.

(2) A one-dimensional array M storing contour data extracted from measurement data as described above. Difference X twice and detect the inflection point by detecting the pixel corresponding to the element that is not "0", set the number of the element of MoX as the Y coordinate, the content of the element of M, X as the X coordinate, Substitute it into the equation of the line segment connecting the detected inflection points and check whether the equation holds. M
Similarly, inflection points are detected for ay, and M is added to the equation connecting the inflection points. Substitute the element of Y and its number and check whether it holds true. The most M among the detected inflection points. Let the inflection point connected by the equation that holds for the elements of X, M, , be one point on the edge of the security.

As a result, even if the security is torn or the edges are bent, one point on the side of the security is detected, the slope θ from the standard position of the security is calculated using this point as a reference, and this slope θ is calculated as a coordinate. By performing the conversion, it is possible to correct the data and compare it with the measured data of the genuine note that falls at the reference position.

The contours of the paper sheets extracted by detecting the points of change in gradation in each row or column are differentiated twice, and ""
Although we have considered a case in which a pixel corresponding to an element that is not O゛'' is an inflection point, it can be applied as long as the inflection point is detected by subtracting the contour of the extracted paper sheet twice.Also,
It is possible to process measured values not only from a two-dimensional image sensor but also from various types of sensors.

In addition, in "1. Correspondence between Examples and FIG. 1",
Although the correspondence between the present invention and the embodiments has been described, those skilled in the art can easily imagine that the present invention is not limited to this and that there are various modifications.

(Effects of the Invention) As described above, according to the present invention, one point on the side of the paper sheet is detected from the measurement data, and the inclination of the paper sheet is calculated using this point as a reference. By performing coordinate transformation, the measurement data is rearranged and the inclination is corrected, and data equivalent to that obtained when the paper sheet is measured at the reference position can be obtained.

This makes it possible to constantly compare the data with the genuine note data at the reference position, and the storage capacity for storing the genuine note data can be extremely reduced, which is extremely useful in practice.

[Brief explanation of the drawing]

FIG. 1 (A) is a principle diagram of the reference point determination method of the first invention, and FIG. 1 (B) is a paper sheet [i 4% system J,
2V is a block diagram of the structure of the securities identification system J according to an embodiment of the present invention. FIG. 3 is a basic block diagram of the embodiment shown in FIG. Fig. 4 is an explanatory diagram of measurement data of securities, Fig. 5 is a configuration V of a conventional securities identification system, and Fig. 6 is an explanatory diagram of the principle of coordinate transformation.In the figure, 104δ:L measuring means, 105 a memory Child actor, 106 is a processing means, 107 +; I discrimination section, 17] is a genuine bill data recording means, 172 is a comparison means, 200.500 is a measurement section, 201.501 is an innocent sensor, 202.502 is an analog-digital (A /D) converter, 203.503 is 1 fastest part, 210.510 is central processing unit (CP [, J), 2
11 is a one-time difference processing unit, 212 is a two-time difference processing unit, 220.520 is written if: jM(, 221, 521 is a measurement data storage area, 222 +, J
Conversion data storage 'In area, 223 is ring γ1; data storage area, 224 +; r, 1 time difference data storage area, 225 is 2 time difference data storage 11 (area, 230.530 is discrimination section, 231.531 is the genuine bill data storage area, 232.532
is a comparison unit, and 240.5/IO is an input/output device. 2. Principle diagram of the first invention Figure 1 (A) 1986 8J] 3+1 Director General of the Japan Patent Office Yoshi 1) Moon Tsuyoshi 1, Indication of matter A9 Patent Application No. 105817 1988 2, Title of the invention Reference point determination method and paper sheet identification system using it 3, relationship with amendment cases Patent 11) J1 Address 1015 I-in 1 Naka/Kami, Nakahara-ku, Yozaki City, Kanagawa Prefecture Name (5
22) Tomi” 151 Notice (03) 375-1631 of Neto 4 Agent, Address: 2-11-2 Yoyogi, Shibuya-ku, Tokyo

Claims (2)

[Claims] (1) Extract the outline of the paper sheet, perform subtraction twice on the outline to detect inflection points, and match the outline with line segments connecting the detected inflection points. detect a point, identify a point on a side of the paper sheet based on the number of detected matching points, and use the point on the side as a reference point for calculating the inclination of the paper sheet. Reference point determination method. (2) Measuring means (104) for measuring desired physical quantities in each minute region of paper sheets; and storage means (105) for storing the measured values of the physical quantities of paper sheets measured by the measuring means (104). Then, a reference point for calculating the inclination of the paper sheet is determined from the measured value stored in the storage means (105), a necessary amount of correction for the inclination is calculated, and the amount of correction is calculated and stored in the storage means (105). processing means (106) for rearranging the measured values and performing coordinate transformation; and recording genuine bill data which is the result of measuring a genuine bill of the same type as the paper sheet at a certain reference position by the measuring means (104). a genuine note data recording means (171), a comparison means (172) for comparing the genuine note data recorded in the genuine note data recording means (171) and the transformed data coordinate-transformed by the processing means (106); A paper sheet discrimination system comprising: a discrimination section (107) for discriminating paper sheets;