Human face data sortability feature extraction method based on weighted maximum spacing criterion

Abstract

The invention discloses a human face data sortability feature extraction method based on weighted maximum spacing criterion. The problems of poor generalization ability of features extracted by MMC and affected classification accuracy, which are caused by the fact that the existing maximum spacing criterion MMC feature extraction uses limited training samples and cannot accurately estimate true distribution of ultra-high-dimensional space samples, are solved. The method comprises the steps that 1) the inter-class distribution matrix Sb and the intra-class distribution matrix Sw of original data are calculated; 2) inter-class and intra-class distribution matrixes are weighted to acquire the weighted maximum spacing criterion WMMC function; 3) the WMMC criterion function is maximized to acquire a mapping matrix; and 4) the original data are mapped to a WMMC subspace; and 5) classifying is carried out in the WMMC subspace. According to the invention, the method can extract the sortability feature with strong generalization ability under the condition of the ultra-high-dimensional small samples; the human face recognition rate is improved; and the method can be used for the sortability feature extraction of the ultra-high-dimensional small sample data.

Description

technical field [0001] The invention belongs to the technical field of image processing, and further relates to a method for extracting separable features of face data, which can be used for extracting separable features of ultra-high-dimensional small-sample data. Background technique [0002] In recent years, face recognition technology is a research hotspot. Compared with ordinary data, face data has the characteristics of ultra-high-dimensional and small samples. Because the dimensionality of face data is too high, which is equal to the number of image pixels, often tens of thousands of dimensions; while the number of samples is often dozens, compared with the low dimensionality, it is difficult to process face data. Usually, before using face data for classification, it is necessary to perform feature extraction processing on the original data to reduce the dimensionality and improve the efficiency of face recognition. Therefore, there are more and more feature extract...

AI Technical Summary

Problems solved by technology

[0006] However, the shortcomings of the MMC-based face data separability feature extraction method are: the dimension of face data is as high as several thousand or even tens of thousands of dimensions, and the number of samples is often only dozens. When using face data features, the distribution of data is only estimated based on the current limited training samples. Due to the large difference between the number of samples and the dimension of data, it is impossible to accurately estimate the true distribution of samples in ultra-high-dimensional space with very limited training samples, resulting in The features extracted by MMC have poor generalization ability, which ultimately affects the accuracy of classification.

Images