Face recognition method based on separation degree difference supervised locality preserving projection

Abstract

The invention relates to the field of biological feature identification, in particular to a face recognition method based on separation degree difference supervised locality preserving projection (SLPP). The method comprises the following steps of reading face images from face databases, conducting feature extraction on the face images to form face features, conducting the feature extraction on face region images to obtain a transformation matrix required by the feature extraction and features of training face images, conducting the feature extraction on testing face images, and classifying and recognizing through a nearest neighbor classifier based on an Euclidean distance. The method solves the problem of a small sample in face recognition, and allows an SLPP method not to be restricted by the feature dimension reserved in a PCA (Principal Component Analysis) process. The method solves the problems that the small sample results in a singular intra-class separation degree matrix, and the optimal matching dimension of PCA and the SLPP is difficult to select.

Description

technical field [0001] The invention relates to the field of biological feature identification, in particular to a face recognition method based on a supervised part-preserving projection based on separation degree difference. Background technique [0002] Locality Preserving Projection (LPP) is a local linear feature extraction method. As a linear approximation of the Laplacian feature map, it can extract low-dimensional features that reflect the nonlinear manifold of high-dimensional samples, and can also process Data outside the training sample. Locality preserving projection is essentially a kind of unsupervised dimensionality reduction, which cannot make full use of the category information of training samples (He X, Niyogi P.Locality preserving projections[J].Advances in Neural Information Processing Systems.2003,16:153-160 ). [0003] Shen Zhonghua et al. proposed a supervised locality preserving projection method (Supervised Locality Preserving Projection, SLPP) fr...

AI Technical Summary

Problems solved by technology

Locality preserving projection is essentially a kind of unsupervised dimensionality reduction, which cannot make full use of the category information of training samples (He X, Niyogi P.Locality preserving projections[J].Advances in Neural Information Processing Systems.2003,16:153-160 )

[0003] Shen Zhonghua et al. proposed a supervised locality preserving projection method (Supervised Locality Preserving Projection, SLPP) from the perspective of preserving the local structure within the class and the degree of separation between classes, which improved the performance of the locality preserving projection method to a certain extent. However, the objective function determined by this method is in the form of Rayleigh quotient, and its solution process is similar to the traditional Fisher criterion. Therefore, in small-sample applications such as face recognition, the problem of singularity of the intra-class separation matrix will be encountered. The general solution It is a combination of PCA and SLPP, but the dimensionality of features retained by SLPP in the dimension reduction process will be severely limited by the dimensionality of features retained in the PCA process (Shen Zhonghua, Pan Yonghui, Wang Shitong. Supervised local retention projection dimensionality reduction algorithm [J], Pattern Recognition and Artificial Intelligence, 2008,21(2):233-239.)

Images