A Second Generation ID Card Recognition Method Based on Feature Fusion

Abstract

The invention discloses a second-generation ID card recognition method based on feature fusion, reads the face image ID on the second-generation ID card, and captures M face images C on the spot at the same time i (i=1,...,M); For the collected face image ID and face image C i (i=1,...,M) perform face capture respectively, and further perform feature extraction, the feature extraction includes global feature extraction and local feature extraction; the global feature and local feature are calculated similarly in the public feature space to get the similarity P of global features A Similarity to local features P B ; only when P A >α and P B >β, the holder of the second-generation ID card is identified as himself. The present invention solves the problem of poor recognition rate caused by the low quality of second-generation ID card pictures by fusing global features and local features; the present invention uses global features and local features to calculate similarity in the public feature library, The application of face recognition technology has a certain impetus; the present invention has significantly improved the recognition rate than the traditional feature fusion method.

Description

technical field [0001] The invention relates to a face recognition technology of a second-generation ID card, in particular to a recognition method of a second-generation ID card based on feature fusion. Background technique [0002] Face recognition is a technology that uses computers to analyze face images and extract effective identification information from them to identify identities. The research on face recognition began in the middle and late 1960s, and has become a popular research topic in recent years. The reason why face recognition has been paid attention to is because of its important research significance, which is highlighted in two aspects: its contribution to the development of disciplines and its huge potential application prospects: (1) face recognition research can greatly promote multi-disciplinary research. (2) Face recognition as a biometric technology has huge potential application prospects. Compared with fingerprints, voices and other biometric r...

AI Technical Summary

Problems solved by technology

[0007] Since the photos stored inside the second-generation ID card are very small, only 102×126 pixels, and the distance between the two eyes is less than 30 pixels, the traditional face recognition technology has a low recognition rate for such small-sized photos. How to comprehensively utilize the global Features and local features to carry out face recognition of the second generation ID card has a high application value

Images