Face recognition method based on intra-class average maximum likelihood cooperative expressions

Abstract

The invention discloses a face recognition method based on intra-class average maximum-likelihood cooperative expressions. Samples similar to a test sample are selected from each class of training samples to form a neighbor sample set; cooperative expression and reconstruction are carried out on intra-class average images of each class of neighbor samples; and classes of neighbor cooperative expressions of maximum likelihood are selected according to reconstruction errors of the different intra-class neighbor average images. Thus, cooperative expression, aimed at reducing a target class, of the neighbor samples reduces the computational complexity to certain extent, and further improves the recognition rate.

Description

technical field [0001] The invention relates to a face recognition method, in particular to a face recognition method based on the maximum similarity cooperative representation of the average value within a class. Background technique [0002] In the field of computer vision system and pattern recognition, feature extraction and classification are the main problems of face recognition research. Among them, traditional linear algorithms based on feature extraction, such as principal component analysis (PCA), linear discriminant analysis (LDA), and some improved algorithms, aim to extract effective low-dimensional global features that can reflect images from high-dimensional spaces. , for classification. Although it has a good effect in practical applications, the linear algorithm cannot extract the local nonlinear features of high-dimensional images very well. For this reason, many popular learning methods have been produced, such as Local Preserving Projection (LPP), Bounda...

AI Technical Summary

Problems solved by technology

Based on the entire training set, LI selects samples that are more similar to the test samples, which greatly reduces the running time. However, this method is susceptible to image noise interference. When the illumination, posture, angle, etc. change greatly, the recognition effect is average.

On this basis, YIN proposed an improvement scheme, that is, on the training set, by comparing the similarity of all training samples and test sample coefficients, select some samples for local collaborative representation, and the recognition is more robust, but this method passes the sample coefficient The similarity selects the nearest neighbor. Since the coefficient calculation of each sample is based on the entire training set, if the number of training samples increases, the computational complexity will increase, so the actual effect of the algorithm is generally

Images