A heterogeneous iris classifier design method for quickly filtering out fake irises in the preprocessing of iris recognition

Abstract

The invention discloses a heterogeneous iris classifier design method for quickly filtering out fake iris in the iris recognition preprocessing process. The design of the classifier includes an improved fake iris database design, iris image evaluation based on MSE analysis, and SNR-based Analytical iris image evaluation, edge detection, iris image evaluation workflow based on improved edge sharpness analysis. In the field of iris recognition, an iris image evaluation system that combines bilateral filtering, MSE analysis, and SNR analysis is innovatively adopted to target irrelevant categories of iris attacks, defocused iris images, motion blurred iris images, and grayscale features and noise of color contact lenses. The distribution feature realizes its efficient and stable filtering. Further innovatively design the iris image evaluation system based on the improved edge sharpness analysis for the edge sharpness feature of the synthetic iris with high simulation degree. Finally, the rapid and stable filtering of fake irises in the iris recognition preprocessing process is realized.

Description

technical field [0001] The invention relates to the field of computer images, in particular to the CAISA iris database of the Institute of Automation of the Chinese Academy of Sciences and C++ image processing functions, specified grayscale extraction in images, noise reduction and visual display of processing results. Background technique [0002] With the rapid development of network and information technology, information security issues have become increasingly prominent. The improvement of software and hardware technology has enabled the popularization and use of computers and intelligent electronic devices, and accelerated the penetration of network technology into all aspects of social life. Illegal acquisition of users' personal information through network technology happens from time to time, which is hard to prevent, which makes people pay more attention to the security of personal information. Traditional identity verification methods are based on passwords and c...

AI Technical Summary

Problems solved by technology

[0013] The purpose of the present invention is to solve the problem that the traditional heterogeneous iris classifier based on iris texture features needs to perform the filtering operation of heterogeneous iris in the feature extraction link, so that the entire iris preprocessing, positioning and segmentation and normalization operation links are almost invalid. Therefore, a large amount of computing resources are wasted and the overall execution efficiency of the iris recognition system is also reduced.

Images