Face Recognition Based on the Improved Sparse Representation

Abstract

In recent years, face recognition is a hot research topic in the field of biometrics, and sparse representation is a hot topic in this field. In order to solve the problem that the sparse representation algorithm does not have good recognition effect in the face image training samples and test samples including both expression changes and illumination condition changes, a face recognition algorithm based on sparse coefficient and residual fusion is proposed.. The algorithm combines the sparse coefficient into the residual classification decision, considering that the sparse coefficient is the fidelity condition that discriminates the correlation between the feature correlation and the class in the training sample class, and the feature correlation between the training sample and the test sample class. The new residual is defined as the classification standard, and the improved residual fully exploits the correlation information of the sparse coefficient. At the same time, the algorithm in this article is combined with the PCA feature extraction method to effectively reduce the data dimension while ensuring the recognition rate of face images. Several experiments were carried out in the standard face database to verify the effectiveness of the proposed method, and it is proved that the algorithm is robust.