Double mode bio-identifying system based on human face and hand

Abstract

In the system, two matching results of human face and hand shape can be obtained through image preprocesisng and pattern identifying technique separately after image of the human face and the hand shape is collected. The two results are normalized into the same range by utilizing standard normalization method and the decision blending for two obtained results are carried on by utilizing support vector computer method in statistic theory so that the final authentication result can be obtained as the judgement is worked out. The identification result of hand shape has realy a high accuracy and rubustness owing to that the degree of freedom for collecting hand shape image is increased in its identification algorithm.

Description

1. Technical field [0001] The invention belongs to the technical field of pattern recognition, and mainly relates to a dual-mode biometric authentication system based on face recognition and hand shape recognition. 2. Background technology [0002] Biometrics technology refers to a technology that uses human's own physiological or behavioral characteristics for identity authentication through computers. Signature, voice, gait, etc.) as the basis, using computer technology for image processing and pattern recognition for identity authentication. This technology has very good security and reliability. Compared with traditional identity authentication methods (such as passwords, IC cards and ID cards, etc.), it will not be lost and forgotten, and it is difficult to forge and counterfeit. Convenient and secure identity authentication method. [0003] Each biometric technology has its inherent defects: for example, about 4% of human fingerprint images are inferior, and fingerpr...

AI Technical Summary

Problems solved by technology

[0003] Each biometric technology has its inherent defects: for example, about 4% of human fingerprint images are inferior, and fingerprint features are easily affected by noise, so some of the actually collected fingerprint images are difficult to be correctly identified; iris features Susceptible to eye diseases, it is difficult for the general public to accept the collection of iris images, and the collection equipment is relatively expensive; the voice features are also easily affected by the physical condition and environment of the person being recognized; the static features of the signature are easy to be imitated, and the dynamic features are unstable and difficult to extract; Due to the complex technical conditions of face recognition, it has not yet reached the practical level; the uniqueness of hand shape features is poor

The former can only be estimated more accurately when the number of samples is sufficient, while the latter two are not easy to obtain in practical applications, and are often determined based on historical data or experience; and methods based on neural networks also encounter problems in how to determine the network structure. , over-learning, under-learning and falling into local minimum

Images