A method and system for eliminating magnetic point noise on a banknote press shaft

Abstract

The invention discloses a method and system for eliminating magnetic point noise on a banknote pressing shaft. The method includes: acquiring a magnetic data image of banknotes; performing region segmentation on the magnetic data image to obtain a non-magnetic area; obtaining suspicious magnetic blocks in the non-magnetic area; determining that the suspicious magnetic blocks are distributed in a preset period, And the area of ​​the suspicious magnetic block is less than or equal to a preset threshold, and the suspicious magnetic block is filtered. The present invention provides a method for eliminating magnetic point noise on the banknote pressing shaft. By judging the periodic distribution and area characteristics of the doubtful magnetic block, it is determined that the doubtful magnetic block is caused by the magnetic point on the banknote pressing shaft. Filtering, so as to eliminate the influence of magnetic point noise on the banknote pressing shaft on the recognition accuracy of the magnetic algorithm, and improve the accuracy of banknote inspection.

Description

technical field [0001] The invention relates to the field of data processing, in particular to a method and system for eliminating magnetic point noise on a banknote verification shaft. Background technique [0002] In the data environment collected by the magnetic data sensor, when the authenticity of the banknote is identified according to the difference between the magnetic point of the detected banknote at the magnetic point of the magnetic sensor and the magnetic data of the real banknote, due to the influence of various external environments, the magnetic data will have noise. Since the gap between the sensor and the banknote press shaft is generally 0.15-0.3mm, in the banknote data collected by the magnetic sensor, due to the long-term running of banknotes or the influence of high and low temperature, the banknote press shaft will inevitably be weakly magnetized and will be contaminated with magnetism. , so there will be periodic noise introduced by the banknote press...

AI Technical Summary

Problems solved by technology

If such noise is not dealt with, it will seriously affect the accuracy of magnetic algorithm identification

[0003] Among the existing magnetic algorithms, the Fourier transform is used to remove the low-frequency background noise in the magnetically collected data, but part of the periodic noise introduced by the banknote pressing axis is mixed in the high-frequency part, and the effect of low-frequency filtering and denoising is not obvious, and Due to the improvement of sensor resolution, it is not allowed to perform Fourier transform on the entire magnetic data in terms of timeliness, so the current algorithm cannot effectively remove the periodic noise introduced by the banknote pressing axis

Images