JP3423134B2 - Paper sheet identification method - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for identifying paper sheets such as banknotes and securities, and more particularly to an identification method for determining the authenticity of paper sheets.

[0002]

2. Description of the Related Art Conventionally, as this type of paper sheet identification method,
Banknote discrimination provided with a sensor unit for reading a print pattern of a discriminated bill as in Japanese Patent Laid-Open No. 5-143829 and a data storage memory for storing discriminated data including print pattern data of the bill obtained from the sensor unit. In the apparatus, a print pattern data selection unit that retrieves print pattern data of the banknote from the determined data and extracts the print pattern data,
A banknote that calculates print pattern data in the longitudinal direction of the banknote from the print pattern data selected by the print pattern data selecting means, and compares and compares the calculated pattern data with standard data for each denomination direction set in advance. There is a bill discriminating device including a discriminating means.

Further, as in Japanese Patent Publication No. 7-89386, the authenticity of the paper sheet is determined by comparing a detection pattern obtained by detecting a desired portion on the paper sheet with a sensor with a reference pattern. A true / false determination device for a paper sheet, which determines whether or not the detection pattern is included in the reference pattern for each predetermined section of the desired position on the paper sheet, and based on the determination result, A predetermined evaluation point is given for each division, and a predetermined calculation is performed on the evaluation points over the entire desired portion to obtain a matching rate that is the degree of matching with the detection pattern reference pattern, and the matching rate is a predetermined value. In the paper sheet authenticity determination device for determining the paper sheet as true when the above, in the reference pattern,
An authenticity determination device for a paper sheet, which comprises a plurality of basic patterns divided according to the distribution degree of a pattern of a large number of paper sheets, and the evaluation point is different depending on which basic pattern area the detection pattern is included in There is.

[0004]

The former of the prior arts, however, has a problem that the deviation at the time of input cannot be accurately corrected if there is dirt or the like near the leading data of the bill to be identified. .

In the latter case, the data of the desired portion obtained by the sensor must be compared with the reference pattern one by one, resulting in a huge amount of calculation and a long identification time.

The present invention has been made in view of the above problems of the prior art.
An object of the present invention is to accurately calculate the deviation width at the time of inputting a paper sheet with a small amount of calculation and improve the identification accuracy at high speed.

[0007]

SUMMARY OF THE INVENTION The present invention has been made in view of the above problems of the prior art, and compares a detection waveform obtained by detecting a desired region of a paper sheet with a sensor with a reference waveform. In the paper sheet identification method for determining the authenticity of the paper sheet by the above, the detection waveform is shifted by varying the deviation width of the detection waveform with respect to the reference waveform between the minimum deviation width and the maximum deviation width in which the deviation can occur. A plurality of representative waveforms are created, a difference value between the representative waveform and a reference waveform at a plurality of predetermined positions of the representative waveform is calculated, and a deviation width of the representative waveform having a minimum cumulative value of the obtained difference values is calculated. , The deviation width in the sensor detection of the detected paper sheet, and the selection of the plurality of locations, the basic representative waveform is created from the paper sheet without stains or tears, and the basic representative waveform is shifted Create multiple waveforms for both waveforms Calculating the absolute value of the difference value of each position,
It is characterized in that the minimum absolute value of the difference value for each position is obtained, and a plurality of locations are selected in order from the largest absolute value.

The cumulative value of the difference between the representative waveform and the reference waveform is preferably the sum of absolute values of the differences or the sum of squares of the differences.

[0009]

BEST MODE FOR CARRYING OUT THE INVENTION An embodiment of the paper sheet discriminating method of the present invention applied to banknotes will be described in detail below with reference to the drawings.

FIG. 1 is a flow chart showing an authenticity / counterfeit discrimination algorithm as a whole of the bill discriminating method. In the figure, the deviation width at the time of input is calculated from the input data of the identified bill input from the magnetic or optical sensor (step S1), and the input data is corrected with the calculated deviation width (step S2).

Next, the difference between the corrected input data and the reference waveform obtained from the input of the genuine bill sensor in advance is taken to extract the dirt and distortion components of the input data (step S3).

On the other hand, using a plurality of pieces of stain component data obtained in advance from a genuine note, the stain component of the current input data is predicted by a method such as a multiple regression model, an autoregressive model or a neural network (step S4). .

Finally, the difference between the dirt and distortion components of the input data calculated in this way and the prediction data is calculated, and the one whose difference is within a predetermined value is a genuine note.
A bill that exceeds a predetermined value is determined to be a fake bill (step S
5).

The waveform diagram shown in the right half of FIG. 1 is a waveform diagram obtained at each step. Step S
2, when the deviation width is not corrected, as shown in the left half of FIG. 2, there is almost no deviation when the banknote is accurately conveyed, and dirt and distortion components are extracted by calculating the difference from the reference waveform. Is performed with high accuracy, and there is almost no effect on each subsequent step.

On the other hand, as shown in the right half of FIG. 2, when the conveyance deviation occurs, an error occurs in the difference calculation with the reference waveform depending on the deviation width, and the bill to be identified becomes dirty. The distortion component cannot be accurately extracted, and the stain component cannot be predicted.

Next, a method of calculating the transport deviation will be described with reference to the flowchart of FIG. Deviation width K in step S11
Set (for calculation). The deviation width K is a value between the minimum deviation width value and the maximum deviation width value in which the conveyance deviation can occur, and starts from the minimum value.

In step S12, data is generated by shifting the input signal of the bill conveyed by K, which is set.
In step S13, the data at a plurality of target locations are extracted from the data generated in step S12 and shifted by K.

The total absolute value of the differences between the data extracted in step S14 and the corresponding data of the reference waveform is calculated. If the cumulative value obtained in step S15 is the minimum, the value of K at that time is temporarily recorded as the calculated shift width.

The above operation is repeated from the minimum value to the maximum value of K. By doing so, every time the value of K changes from the minimum value to the maximum value, the cumulative value of the difference from the reference waveform is obtained, and the minimum cumulative value of K is updated each time. Go.

Then, by setting the finally remaining value of K as the target conveyance deviation width, the conveyance deviation width can be accurately obtained. If a higher degree of accuracy is required, instead of calculating the absolute value sum of the differences in step S14, the square sum of the differences is calculated (step S141),
The smallest deviation value may be determined as the target deviation width. If the accuracy is even lower, it is possible to determine the difference having the smallest absolute value as the target deviation width without calculating the cumulative total of the differences.

The extraction of data at a plurality of locations in step S13 is performed by the method shown in FIG. That is, in step S131, a basic representative waveform is created from a bill (completed bill) that is not stained or broken.

In step S132, a basic representative waveform in which the shift width K is set and shifted is created in the same manner as in the previous calculation of the transport shift width. In step S133, the difference value (absolute value) of each position between the shifted basic representative waveform and the original basic representative waveform is recorded.

Then, the above operation is performed to change the value of K from the minimum value to the maximum value and successively record the difference value, and in step S134, the minimum difference value at each position is calculated. Finally, plural points are selected in order from the one having the smallest minimum difference value obtained in step S135, and these points are set as plural points from which data should be extracted.

The conceptual diagram of the waveform in the middle is shown in the right half of FIG.
The left half shows a conceptual diagram of the steps for selecting multiple locations.

[0025]

As described above, the present invention compares the detection of the conveyance deviation width with the reference data by using only the data at a plurality of positions in the input data when the inputted paper sheets are identified. By doing so, it is possible to accurately calculate the conveyance deviation width in a short time by performing it accurately, and it is possible to perform the sheet identification speed as a whole without lowering the accuracy.

[Brief description of drawings]

FIG. 1 is a flowchart showing an overall algorithm for paper sheet identification.

FIG. 2 is a conceptual diagram for explaining the influence of conveyance deviation on sheet identification.

FIG. 3 is a flowchart illustrating a method for calculating a conveyance deviation width.

FIG. 4 is a flowchart illustrating a method of selecting data at a plurality of locations from input data.

─────────────────────────────────────────────────── ─── Continuation of the front page (58) Fields surveyed (Int.Cl. ⁷ , DB name) G07D ^7/ 00-7/20 G06F 17/60-19/00

Claims (2)

(57) [Claims] 1. A paper sheet identification method for determining the authenticity of a paper sheet by comparing a detection waveform obtained by detecting a desired region of a paper sheet with a sensor, to a reference waveform, the reference of the detection waveform. A plurality of representative waveforms in which the detected waveform is shifted by varying the shift width with respect to the waveform between the minimum shift width and the maximum shift width in which the shift can occur are created, and the representative waveforms at a plurality of predetermined locations of the representative waveform are generated. It calculates a difference value between the reference waveform, the deviation of the representative waveform cumulative value is the smallest obtained difference value, along with a shift width of the sensor detection of the paper sheets to be detected, the selection of the plurality of locations Is based on paper sheets that have no stains or tears.
This representative waveform is created and the basic representative waveform is shifted
Create a number of representative waveforms and save the difference value at each position of both waveforms.
The absolute value is calculated, and the absolute value of the difference value for each position is the minimum
Calculate the value and select multiple locations in order from the largest minimum value
A method for identifying a paper sheet, which comprises: 2. A cumulative total of differences between the representative waveform and the reference waveform.
The value is the sum of absolute values of each difference or the sum of squares of each difference.
The paper sheet identification method according to claim 1.