Kernel image differential filter designing method based on learning and characteristic discrimination

Abstract

The invention discloses a kernel image differential filter designing method based on learning and characteristic discrimination. The method includes the steps of stretching an original image block vector corresponding to any point p in a linear discrimination filter in rows about the partial region with p as the center to form an image block vector of neighboring domain pixel information only containing p, conducting dimension expansion on the basis of the image block vector, serially connecting the first and second order differential information at the pixel in the original image block vector, constructing an image block vector matrix to form an intra-class dispersion matrix and inter-class dispersion matrix, and introducing kernel operation for the intra-class dispersion matrix and inter-class dispersion matrix. Filter learning is conducted in a high dimension space in combination with a kernel method, and linear discrimination analysis concept is merged into the learning process, so that detail and non-linear information in an image can be better used to obtain an image filter with better discrimination characteristic description.

Description

technical field [0001] The invention relates to the field of pattern recognition and classification, in particular to a method for designing a nuclear image differential filter based on learning and feature identification, especially for the classification and recognition of human faces or specific targets. Background technique [0002] Extracting efficient and discriminative feature descriptions is a key issue in face recognition and other pattern recognition applications. The quality of feature extraction will directly affect the performance of subsequent classification and recognition. Taking the face recognition application as an example, the goal of feature extraction is to increase the intra-class similarity of features and reduce the inter-class similarity as much as possible while obtaining high-efficiency discrimination. However, affected by various factors such as expression, illumination, pose, occlusion, etc., efficient and robust feature extraction is still a ho...

AI Technical Summary

Problems solved by technology

Since IFL directly performs filter learning in the pixel space, factors such as illumination, expression, and posture often cause nonlinear problems in the face recognition process, and linear IFL cannot handle such problems better.

In addition, in order to simulate the idea of ​​LBP's neighborhood pixel comparison, IFL only uses the first-order differential information of the image in the filter learning process, while the high-order differential information such as corner points and gray-level change rates in the image is not sufficient. use

Images