A method for banknote magnetic anti-counterfeiting features based on the ViT classification model of magnetic spectrum

Abstract

This invention discloses a method for banknote magnetic anti-counterfeiting feature authentication based on a magnetic spectrum ViT classification model, comprising: acquiring banknote magnetic signals; reading the magnetic spectrum ViT classification model, the model having predetermined upper limit of intensity, predetermined lower limit of intensity, predetermined upper limit of frequency, predetermined signal length, and predetermined velocity-frequency ratio data of the magnetic feature signal; detecting and truncating the effective magnetic feature signal in the magnetic signal; adjusting the signal length to a predetermined signal length according to the velocity-frequency ratio of the truncated effective magnetic feature signal; performing magnetic spectrum transformation on the magnetic feature signal according to the velocity-frequency ratio of the magnetic feature signal, and drawing a magnetic spectrum using the predetermined upper limit of intensity, predetermined lower limit of intensity, and predetermined upper limit of frequency of the magnetic feature signal; inputting the drawn magnetic spectrum into the magnetic spectrum ViT classification model for recognition, analyzing the recognition results, and completing the banknote magnetic anti-counterfeiting feature authentication based on the recognition results.

Description

Technical Field

[0001] This invention relates to the field of banknote authentication technology, and in particular to a method for banknote magnetic anti-counterfeiting feature authentication based on the ViT classification model of magnetic spectrum. Background Technology

[0002] With the continuous development of magnetic sensor technology, magnetic image sensors have been successfully developed and are beginning to be applied in the field of banknote authentication. Magnetic image sensors, in conjunction with a uniform-speed paper feed device, can detect the complex magnetic distribution on banknotes and convert it into a magnetic image, thus enabling quantitative authentication of banknotes based on their magnetic characteristics. Therefore, quantitative analysis and authentication of the magnetic distribution on banknotes using magnetic images has significant research value and great application potential in the field of counterfeit detection. Summary of the Invention

[0003] To address the aforementioned technical problems, the purpose of this invention is to provide a method for identifying counterfeit banknotes using magnetic anti-counterfeiting features based on the ViT classification model of magnetic spectrum.

[0004] The objective of this invention is achieved through the following technical solution:

[0005] A method for banknote magnetic security feature authentication based on a magnetic spectrum ViT classification model includes:

[0006] A collects the magnetic signal of banknotes;

[0007] B reads the magnetic spectrum ViT classification model, which has predetermined upper limit of intensity, lower limit of intensity, upper limit of frequency, predetermined signal length, and predetermined velocity-frequency ratio data for the magnetic feature signal; detects and extracts the effective magnetic feature signal in the magnetic signal;

[0008] C adjusts the signal length to a predetermined signal length based on the velocity-frequency ratio of the extracted effective magnetic characteristic signal;

[0009] D performs magnetic spectrum transformation on the magnetic characteristic signal based on the velocity-frequency ratio of the magnetic characteristic signal, and plots the magnetic spectrum using the predetermined upper limit of intensity, the predetermined lower limit of intensity and the predetermined upper limit of frequency of the magnetic characteristic signal.

[0010] E inputs the drawn magnetic spectrum into the magnetic spectrum ViT classification model for identification, analyzes the identification results, and completes the banknote magnetic anti-counterfeiting feature authentication based on the identification results.

[0011] Compared with the prior art, one or more embodiments of the present invention may have the following advantages:

[0012] This method addresses the instability of scanning speed and frequency during actual scanning by preprocessing the signal. Based on the speed-frequency ratio of the magnetic characteristic signal, the signal length is adjusted to a predetermined length to resolve the inconsistency of the horizontal axis in the magnetic spectrum. By using the speed-frequency ratio for each scan, consistency of the vertical axis of the magnetic spectrum can be guaranteed. Fixing the upper and lower intensity limits ensures consistency of the intensity axis (color) of the magnetic spectrum. This comprehensively guarantees the consistency of the magnetic spectrum. Figure 1 Consistency is ensured to guarantee the accuracy of recognition. Attached Figure Description

[0013] Figure 1 This is a flowchart of a banknote magnetic anti-counterfeiting feature authentication method based on the magnetic spectrum ViT classification model. Detailed Implementation

[0014] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be described in further detail below with reference to the embodiments and accompanying drawings.

[0015] like Figure 1 The diagram shows the workflow of a banknote magnetic security feature authentication method based on the ViT classification model of magnetic spectrum, including:

[0016] A collects the magnetic signal of banknotes;

[0017] Along the short side of the banknote, magnetic signals are collected using multiple magnetic sensors at key locations specified on the banknote. These key locations with magnetic anti-counterfeiting features include: the security thread, the two-color irregular horizontal serial number, and the main background pattern.

[0018] B uses the ViT classification model of the magnetic spectrum, which has predetermined upper limit of intensity, lower limit of intensity, upper limit of frequency, predetermined signal length, and predetermined velocity-frequency ratio data for the magnetic feature signal; it detects and extracts the effective magnetic feature signal in the magnetic signal.

[0019] Effective magnetic feature signals in magnetic signals can be detected by learning using Transformer or LSTM deep learning models, or by detecting them based on signal strength.

[0020] For detection based on signal strength, a 1×n-dimensional convolution operator is used. det =[1,1,1,....,1,]; Let the magnetic signal be S. mag The upper limit of the intensity is p signal_upper， Then determine if it has

[0021] Magnetic characteristic signal S mag_on

[0022] S mag_on ={x|x∈S mag *I det x>0.05×p signal_upper}

[0023] C adjusts the signal length to a predetermined signal length based on the velocity-frequency ratio of the extracted effective magnetic characteristic signal;

[0024] In this scan, the average scan speed v signal The magnetic sensor sampling frequency is f signal The banknote is L long. cash That is, the velocity-frequency ratio of the magnetic characteristic signal can be obtained as r. signal Signal length l signal :

[0025] r signal =f signal / v signal

[0026] l signal =L cash ×f signal / v signal

[0027] For the predetermined signal length l signal_set , predetermined speed-frequency ratio r signal_set .

[0028] Then the normalized predetermined signal length l can be obtained. signal_set_norm for

[0029] l signal_set_norm =l signal_set ×r signal ÷r signal_set

[0030] The process of adjusting the signal length to the predetermined length is as follows:

[0031]

[0032] It can solve the problem of consistency of the horizontal axis when drawing magnetic spectrum.

[0033] D performs magnetic spectrum transformation on the magnetic characteristic signal based on the velocity-frequency ratio of the magnetic characteristic signal, and plots the magnetic spectrum using the predetermined upper limit of intensity, the predetermined lower limit of intensity and the predetermined upper limit of frequency of the magnetic characteristic signal.

[0034] The velocity-frequency ratio of the magnetic characteristic signal is r signal The upper limit of the intensity of the predetermined magnetic characteristic signal is p. upper The lower limit of intensity is p lower The upper limit of frequency is f upper Drawing a magnetic spectrum includes:

[0035] Plotting a magnetic spectrum according to the upper and lower limits of intensity and the upper limit of frequency means setting the starting point of the horizontal axis of the magnetic spectrum to 0 and the ending point to the signal length l. signal;

[0036] The vertical axis of the magnetic spectrum starts at 0, and the horizontal axis ends at the upper limit of the frequency value f. upper ;

[0037] The starting point of the intensity color axis of the magnetic spectrum is: the lower limit p of the magnetic sensing signal intensity. down The endpoint of the intensity color axis is the upper limit p of the magnetic sensing signal intensity. upper .

[0038] By employing a frequency ratio for each scan, the consistency of the vertical axis of the magnetic spectrum can be guaranteed. Fixing the upper and lower intensity limits ensures consistency (color) along the intensity axis of the magnetic spectrum. This comprehensively guarantees the consistency of the magnetic spectrum. Figure 1 Consistency is ensured to guarantee the accuracy of recognition.

[0039] E inputs the drawn magnetic spectrum into the magnetic spectrum ViT classification model for identification, analyzes the identification results, and completes the banknote magnetic anti-counterfeiting feature authentication based on the identification results.

[0040] The identification types include: suspicious, security thread, two-color irregular horizontal serial number, main background pattern, and other self-defined magnetic features; the identification results are determined based on the identification types. The analysis method for the identification results is shown in Table 1.

[0041] Table 1

[0042]

[0043] For the identification results, if any item is suspicious, the confidence level is below 90%, or the magnetic feature identification results are inconsistent, then the individual item is set as suspicious and the overall conclusion is suspicious. Only when all scan positions meet the normal requirements is the overall conclusion given as normal.

[0044] While the embodiments disclosed in this invention are as described above, the content is merely for the purpose of facilitating understanding of the invention and is not intended to limit the invention. Any person skilled in the art to which this invention pertains may make any modifications and variations in form and detail of the implementation without departing from the spirit and scope disclosed herein; however, the scope of patent protection for this invention shall still be determined by the scope defined in the appended claims.

Claims

1. A method for identifying counterfeit banknotes based on the ViT classification model of magnetic spectrum, characterized in that, include: A collects magnetic signals from banknotes; B reads the magnetic spectrum ViT classification model, which has predetermined upper limit of intensity, lower limit of intensity, upper limit of frequency, predetermined signal length, and predetermined velocity-frequency ratio data for the magnetic feature signal; detects and extracts the effective magnetic feature signal in the magnetic signal; C adjusts the signal length to a predetermined signal length based on the velocity-frequency ratio of the extracted effective magnetic characteristic signal; D performs magnetic spectrum transformation on the magnetic characteristic signal based on the velocity-frequency ratio of the magnetic characteristic signal, and plots the magnetic spectrum using the predetermined upper limit of intensity, the predetermined lower limit of intensity and the predetermined upper limit of frequency of the magnetic characteristic signal. E inputs the drawn magnetic spectrum into the magnetic spectrum ViT classification model for identification, analyzes the identification results, and completes the banknote magnetic anti-counterfeiting feature authentication based on the identification results; The C includes: During the scan, let the average scan speed be... The magnetic sensor sampling frequency is The length of the banknote is The velocity frequency ratio of the magnetic characteristic signal is... With signal length The calculation formula is: ； The predetermined signal length is The predetermined speed-to-frequency ratio is Through a predetermined signal length of The predetermined speed-to-frequency ratio is and the velocity sampling ratio of magnetic characteristic signals Obtain the normalized predetermined signal length : ； The signal length will then be adjusted to the predetermined length: 。 2. The banknote magnetic anti-counterfeiting feature authentication method based on the magnetic spectrum ViT classification model according to claim 1, characterized in that, The A includes: along the short side of the banknote, at designated key locations on the banknote, using a multi-channel magnetic sensor to collect magnetic signals; the key locations are the security thread, the two-color irregular horizontal serial number, and the main background pattern.

3. The banknote magnetic anti-counterfeiting feature authentication method based on the magnetic spectrum ViT classification model according to claim 1, characterized in that, The effective magnetic feature signals in the magnetic signal in B are detected after being learned by a Transformer or LSTM deep learning model.

4. The banknote magnetic anti-counterfeiting feature authentication method based on the magnetic spectrum ViT classification model according to claim 1, characterized in that, The effective magnetic characteristic signal in the magnetic signal of B is detected based on the signal strength, including: Using convolution operators 1×n dimensional I det =[1,1,1,....,1,];Let the magnetic signal be... S mag The upper limit of intensity is p signal_upper Then the magnetic characteristic signal S mag_on The calculation formula is: 。 5. The banknote magnetic anti-counterfeiting feature authentication method based on the magnetic spectrum ViT classification model according to claim 1, characterized in that, The velocity-frequency ratio of the magnetic feature signal in D is: The upper limit of the predetermined intensity of the magnetic characteristic signal is r The lower limit of the predetermined intensity is The predetermined frequency limit is ; through the velocity-frequency ratio of the magnetic characteristic signal The predetermined intensity upper limit of the magnetic characteristic signal Predicted lower limit of intensity With the predetermined frequency limit Drawing a magnetic spectrum includes: Plotting a magnetic spectrum according to the upper and lower limits of intensity and the upper limit of frequency means setting the starting point of the horizontal axis of the magnetic spectrum to 0 and the ending point to the signal length. ; The vertical axis of the magnetic spectrum starts at 0, and the horizontal axis ends at the upper limit of the frequency value. ; The starting point of the intensity color axis of the magnetic spectrum is the lower limit of the intensity of the magnetic characteristic signal. The endpoint of the intensity color axis is the upper limit of the intensity of the magnetic feature signal. .

6. The banknote magnetic anti-counterfeiting feature authentication method based on the magnetic spectrum ViT classification model according to claim 1, characterized in that, The identification types include: suspicious, security line, two-color irregular horizontal serial number and main background pattern; the identification result is determined based on the identification type.

7. The banknote magnetic anti-counterfeiting feature authentication method based on the magnetic spectrum ViT classification model according to claim 6, characterized in that, If any of the following identification results in the identification type have a confidence level of less than 90%: suspicious or security line, two-color irregular horizontal number, main background pattern, or other self-defined magnetic features, the identification result is set as suspicious.

Images