Neighborhood prior embedded collaborative representation mode identification method

Abstract

The invention provides a neighborhood prior embedded collaborative representation mode recognition method, which is used for solving the technical problems of low accuracy, poor robustness, poor generalization ability and small sample over-fitting of an existing face recognition method. The method comprises the following steps: firstly, preprocessing a face image in a face database to obtain a training sample and a test sample; secondly, obtaining neighbor priori information of the test sample according to the training sample and the test sample; constructing an embedded collaborative representation model based on neighborhood prior according to the neighborhood prior information of the test sample; and finally, solving the embedded collaborative representation model based on neighborhoodprior to obtain an optimal coding coefficient vector, and classifying the test samples according to the optimal coding coefficient vector. According to the method, the importance of the local consistency information of the sample is considered, and the obtained neighbor priori of the test sample is embedded into the original collaborative representation model, so that the classification accuracy and robustness of the classifier are improved, and over-fitting is prevented.

Description

technical field [0001] The invention relates to the technical field of face recognition, in particular to a neighborhood prior embedded collaborative representation pattern recognition method. Background technique [0002] Face recognition has become a popular research topic in the field of computer vision due to its wide application in information security, video surveillance, urban rail transit safety, etc. However, in real life, the real shooting environment such as changing expressions and lighting, different postures and occlusions will reduce the quality of facial images, making face recognition more difficult. In addition, the diversity of requirements and the complexity of application scenarios make it more challenging to design flexible and effective classifiers. Face recognition systems generally use high-quality feature extraction techniques, and the success of the system largely depends on the performance of the classifier. Therefore, designing a powerful class...

AI Technical Summary

Problems solved by technology

[0004] Aiming at the defects in the above-mentioned background technology, the present invention proposes a neighborhood prior embedded collaborative representation pattern recognition method, which solves the problems of low accuracy, poor robustness, poor generalization ability, small The technical problem of sample overfitting

Images