A Face Recognition Method Based on Two-Dimensional Linear Discriminant Analysis

Abstract

The invention discloses a face recognition method of two-dimensional linear discriminant analysis in the field of biological pattern recognition. Affinity matrix elements are calculated according to the selected training sample matrix, and intra-class divergence matrix and class scatter matrix are calculated according to the affinity matrix elements. The inter-class scatter matrix, according to the intra-class scatter matrix and the inter-class scatter matrix, calculates the eigenvalues ​​and eigenvectors of the matrix product of the inverse of the intra-class scatter matrix and the inter-class scatter matrix, and obtains the projection matrix; using projection The matrix projects the training sample matrix to the projection space to obtain the projected matrix: the nearest neighbor classifier is used to classify the projected matrix and the test samples, and the recognition rate is calculated; the present invention combines two-dimensional linear discriminant analysis and two-dimensional local preservation projection The advantage of the method is that it can not only effectively reduce the dimensionality of the original data, but also retain the local characteristics of the data, avoiding the destruction of the original two-dimensional structure of the sample data when it is stretched into one-dimensional data, and has high recognition rate and high efficiency.

Description

technical field [0001] The invention relates to the field of biological pattern recognition, in particular to a face recognition technology. Background technique [0002] With the development of computer science and biomedical technology, the use of human biometrics for identification has become an important way. As an important part of biometric technology, face recognition technology has become the most commonly used means of identification in people's daily life. It not only has the universality, security, uniqueness, stability and collectability At the same time, it has the advantages of no need for target cooperation, long-distance execution and intuitive comparison. Therefore, face recognition technology has been widely used in information security, criminal investigation, entrance and exit monitoring and other fields. [0003] Existing face recognition methods include: Principal Component Analysis (PCA), Linear Discriminant Analysis (LDA), Kernel-based Feature Extra...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to solve the problems existing in the existing linear discrimination analysis method, and propose a face recognition method of two-dimensional linear discrimination analysis, which can directly perform calculation processing on the two-dimensional face data without destroying the face data structure, which simplifies the calculation process and improves the face recognition rate

Images