A method of RMB banknote denomination recognition based on image histogram matching

Abstract

The invention provides an RMB paper currency denomination identification method based on histogram matching of images. The method is applied to RMB paper currency denomination identification. The training process comprises the following steps: acquiring a two-dimensional matrix of image training samples; filtering each image training sample; extracting a first histogram vector of a red-green-blue three-color component of each image training sample; and respectively acquiring the average value of the histogram of each color component in the multiple image training samples to serve as a template of the paper currency with the denomination. The identification process specifically comprises the following steps: acquiring a two-dimensional matrix of to-be-identified images; filtering the to-be-identified images; extracting a second histogram of the red-green-blue three-color component with an identification image, respectively acquiring the distance between the second histogram and the template of the paper currency of each denomination, comparing the distances, and judging the denomination which corresponds to the template of the paper currency with the denomination with the smallest distance away from the histogram as the denomination of the to-be-identified paper currency. According to the method provided by the invention, the noise robustness is guaranteed, the identification precision is greatly improved, and the identification time is short.

Description

technical field [0001] The invention relates to the financial field, in particular to a banknote denomination recognition method based on image histogram matching. Background technique [0002] With the prosperity and development of the economy, the circulation of banknotes is increasing, and the core technologies of the multi-spectrum banknote counters, sorting machines, and ATM machines used by many domestic banks are all from abroad, which are not only expensive but also endanger financial security. Hidden danger. [0003] The core technical basis of multi-spectral banknote counters, sorting machines, and ATMs is real-time banknote image processing and recognition. In China, the image denomination recognition technology adopted by multi-spectral banknote counters mostly uses artificial feature parts. Due to the difficulty of avoiding image offset and limited counterfeit identification features, its stability and recognition ability are difficult to meet the requirements....

AI Technical Summary

Problems solved by technology

In China, the image denomination recognition technology adopted by multi-spectral banknote counters mostly uses artificial feature parts. Due to the difficulty of avoiding image offset and limited counterfeit identification features, its stability and recognition ability are difficult to meet the requirements.

Images