Multi-view subspace clustering method based on block diagonal representation and view consistency

Abstract

The invention relates to the field of computer vision application, and provides a multi-view subspace clustering method based on block diagonal representation and view consistency, which can improve the multi-view data set clustering precision, and comprises the following steps of: constructing a multi-view data matrix according to an input multi-view data set; constructing a shared consistency representation matrix according to the multi-view data matrix to obtain an objective function; transforming the objective function and obtaining an augmented Lagrangian equation of the objective function; obtaining an updated expression of a variable of the Lagrangian equation through an augmented Lagrangian equation; initializing variables, setting iteration conditions or iteration times, performing iteration updating on the variables according to an updating expression, and outputting a shared consistency representation matrix which meets the iteration conditions or the iteration times as an optimal shared consistency representation matrix; and constructing a hypergraph through the optimal sharing consistency representation matrix, and then clustering the multi-view data set by using a spectral clustering method based on the hypergraph to obtain a clustering result of the multi-view data set.

Description

technical field [0001] The present invention relates to the field of computer vision applications, and more specifically, to a block-diagonal representation and visual Figure 1 Consistent multi-view subspace clustering method. Background technique [0002] Clustering is a powerful data analysis and data processing technique that uses the similarity between data to divide it into multiple clusters. From the perspective of machine learning, clustering is an unsupervised learning method that can cluster data with unknown category information to extract useful information. However, existing clustering methods are usually suitable for single-view data. If all views are concatenated into one view, and state-of-the-art clustering methods are used on this single view, there is still a problem of poor clustering performance, because each Each view has its own specific statistical properties, so this treatment is meaningless. In contrast, multi-view clustering can effectively handl...

AI Technical Summary

Problems solved by technology

However, most of the existing multi-view clustering methods are spectral clustering based on ordinary undirected graphs, which can only describe the relationship between pairs of data points, resulting in the accuracy of multi-view clustering It cannot meet the needs o...

Images