Sparse coding method based on local and global regularization

Abstract

The invention discloses a sparse coding method based on local and global regularization, pertaining to the technical field of sparse coding (SC). The method utilizes local regression regularization tofind out a potential geometric structure of data. Concretely speaking, the overall data space is divided into a large number of partial regions. Each sample is linearly represented by a sample of a partial region to which each sample belongs, which is called as a local learning conception. On the basis of the local learning conception, a kernel global regression method is utilized for grabbing aglobal geometric structure of data. Therefore, partial and global regularization are utilized for obtaining an inherent geometric structure of data.

Description

technical field [0001] The invention belongs to the technical field of sparse coding (SC), in particular to a sparse coding method based on local and global regularization. Background technique [0002] In many classification and clustering problems, the processing of high-dimensional data is a very challenging problem. In order to solve this problem, people often use data representation methods to find low-dimensional representations in high-dimensional data, so as to achieve the purpose of improving computing efficiency and reducing storage space. At present, data representation methods have attracted the attention of many scholars due to their excellent representation in computer vision, information retrieval and machine learning. [0003] Principal component analysis (PCA) and linear discriminant analysis (LDA) are two popular linear representation methods, the former is an unsupervised learning method whose purpose is to find the minimum covariance projection direction...

AI Technical Summary

Problems solved by technology

[0021] In order to solve the defect that the low-dimensional representation in the prior art cannot maintain the local and global geometric structure of the data at the same time, the present invention provides a sparse coding method based on local and global regularization. Compared with the traditional sparse coding method, the method constructs Local regression regularization to maintain the local structure information of the data, and kernel regression to capture the global geometric structure of the data

Images