High anti-spoofing multi-modal identity verification method based on competition and cooperation mechanism

Abstract

The invention discloses a high anti-spoofing multi-modal identity verification method based on a competition and cooperation mechanism, belongs to the mode identification field, and comprises the following steps: S1, for each modal biological characteristic, using a characteristic template of a target class and a competition class to form a verification dictionary, carrying out sparse representation on an input sample characteristic, and calculating a matching degree and a correlation degree of an input sample and the target class; S2, evaluating the quality of each modal input sample, and calculating an adaptive coding weight according to the quality; S3, performing multi-modal adaptive joint sparse representation on the input sample features, and calculating a multi-modal matching degreeand relevancy; and S4, cascading the multi-modals and the matching degrees and relevancy of the modals to form a multi-modal matching vector, and inputting the multi-modal matching vector into a classifier for identity verification. Before the system is deployed, theclassifier is trained by using a multi-modal matching vector positive sample, a common negative sample and a spoofing sample set, sothat the classifier has the capability of identifying the common negative sample and the spoofing sample, the spoofing resistance of multi-modal identity verification is effectively improved, and meanwhile, the verification precision is not reduced.

Description

technical field [0001] The invention belongs to the field of biometric identity verification, and mainly relates to pattern recognition technology, in particular to a highly anti-spoofing multi-mode identity verification method based on a competition and cooperation mechanism. Background technique [0002] Spoofing attacks use fake biometrics on the sensor side, such as silicone fingerprints, face photos, videos, 3D masks or head models, etc. to fool biometric authentication systems. If the single-modal authentication system cannot effectively verify that the collected biometric information is from a fake object, the possibility of being cracked is very high. Multimodal biometrics are often considered highly secure because they require users to enter multiple types of biometrics. However, research in recent years has shown that when one or several biometric features of users are forged and used in spoofing attacks, the possibility of successful spoofing of multimodal system...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to provide a highly anti-spoofing multi-modal identity verification method based on a competition and cooperation mechanism to solve the above-mentioned problem of greatly reducing the verification accuracy in the case of ordinary verification without significantly improving the anti-spoofing performance of the system question

Images