Identity recognition method based on face-fingerprint cooperation

Abstract

The invention discloses an identity recognition method based on face-fingerprint cooperation and belongs to the field of machine vision and pattern recognition utilization. The identity recognition method based on the face-finger cooperation is characterized in that the capacity that the human can cognize the world by comprehensively utilizing the internal relations between various information and heterogeneous information is simulated to try to find the relevance between face characteristics and fingerprint characteristics, and multi-modal biological characteristic identification is performed on a self-constructed face-fingerprint cooperation database by utilizing faces, fingerprints and the relevance between the faces and the fingerprints. The identity recognition method based on the face-fingerprint cooperation comprises the following specific steps of: 1, selecting and fusing low-level visual features of an image; 2, expressing high-level semantic features of the image; 3, analyzing the relevance between the face characteristics and the fingerprint characteristics; 4, performing multi-modal information interaction of the high-level semantic features; and 5, establishing a face-fingerprint cooperation database. Verification is performed by the method on the large-sized database containing the face and the fingerprints of the same person. The identity recognition method based on the face-fingerprint cooperation is higher in accuracy and stronger in robustness.

Description

technical field [0001] The present invention relates to a technology utilizing machine vision and pattern recognition, and more specifically, relates to an identity recognition method based on face-fingerprint collaboration. Background technique [0002] Biometric identification technology can provide a convenient, safe, and reliable identity authentication solution, and it is a hot issue in the current academic fields such as information security, image processing, and pattern recognition. However, since the research on biometric identification technology is still in the initial stage of development, how to improve the accuracy and robustness of the identification system is also a difficult issue of great concern in academic fields such as information security, image processing, and pattern recognition. In recent years, the research on single-modal biometric identification technologies such as face recognition and fingerprint recognition has made great progress. At present,...

AI Technical Summary

Problems solved by technology

This assumption ignores the internal connection between the various biometric characteristics of the same person, so it cannot fundamentally improve the performance of biometric identification technology, thus making the development of multimodal biometric identification technology in a bottleneck state.

Images