Facial feature cryptographic key generation-based internet identity authentication method

Abstract

The invention discloses a facial featured cryptographic key generation-based internet identity authentication method, which relates to the technical fields of biological feature recognition and authentication and password application. The method comprises an identity registration stage and an identity authentication stage. In the identity registration stage, a stable cryptographic key is generated by utilizing a facial feature method based on random projection and statistical optimum, stored into a database and used for performing identity authentication on a user. In the identity authentication stage, the steps of feature extraction, feature random projection, cryptographic key generation and feature error correction are performed on a face image by adopting a method consistent with the method for generating the cryptographic key in the identity registration stage, and then a feature matching operation is performed to realize the identity authentication of the user. By the method, a face template is effectively protected from attacks in a face recognition process, good references and choices are provided for the generation, management and the like of the cryptographic key required by an identity authentication system in real life, and the problem of internal diversity in biological feature recognition is effectively solved; and the method is simple and easy.

Description

technical field [0001] The invention relates to the technical field of biometric identification authentication and password application, in particular to the research and realization of an online identity authentication method based on facial feature key generation. Background technique [0002] Biometric identification can effectively solve the problems of user identification and security control in the identity management system. Biological characteristics include physiological characteristics and behavioral characteristics. Commonly used physiological characteristics include fingerprints, hand shape, face shape, iris, retina, and smell; behavioral characteristics include keystrokes, voice, handwriting, gait, etc. With the wide application of biometric systems in many occasions, people are more and more concerned about the security and privacy of biometric technology. An ideal biometric system should have the characteristics of robustness, low error rate and not easy to ...

AI Technical Summary

Problems solved by technology

[0012] There are many deficiencies in the two methods mentioned above, which are mainly reflected in: (1) Different keys are generated according to different avatars of the same person, and there is a large gap between the keys, that is, the intra-class difference of biological characteristics is relatively small. large, which will lead to an increased spurious rejection rate

Images