One-to-one iris identification method for scale variation stability characteristics and multi-algorithm voting

Abstract

The invention discloses a one-to-one iris identification method for scale variation stability characteristics and multi-algorithm voting. Personal information relates to an iris and identification people is stored into an infrared radio frequency scanning identity card to serve as a comparison party in an experiment, and then, the iris of a testing person is collected during normal collection to serve as a testing party in the comparison experiment. By use of the method, on a premise that the effective characteristics of the iris are guaranteed, the interference of redundancy and noise is reduced, a vote is taken through multiple algorithms, a risk of an identification error situation due to a collection environment influence is lowered, the correctness identification rate and the robustness of the iris can be increased, safety and reliability are increased, and the method is simple in operation and easy in mastering.

Description

technical field [0001] The invention relates to the fields of electricity, digital image processing, digital signal transmission and biological feature identification. The concept of scale-varying stable features is proposed. And extract the information of the scale change stable characteristics of the collected iris, and use the method of multi-algorithm voting to determine whether two irises are of the same category in a one-to-one form. Background technique [0002] At present, biometric technology is widely used. Face recognition, iris recognition, fingerprint recognition, etc. have begun to appear in our daily life in large numbers. For some places that require high confidentiality and high security, such as prisons, banks, military bases, etc., iris is the most effective and reliable biometric feature. The key to iris recognition is feature extraction and expression. [0003] Iris feature extraction is currently mainly divided into three types of methods: the first ...

AI Technical Summary

Problems solved by technology

[0004] The three types of algorithms have achieved good results in experiments, but there are still some problems. First, the first and second types of algorithms are aimed at pixel texture information and require high image quality, so the ability to resist noise is poor.

The feature extraction of the third type of algorithm is more complex, and is greatly affected by light.

Secondly, the distribution of iris texture features is not stable, so the iris feature extraction effect may be biased

Therefore, facing the same two iris images and making the same judgment, the results of the three types of algorithm judgments may be different, and there is a possibility of misjudgment.

Images