An Unsupervised Object Recognition Approach Combining Multi-Source Feature Learning and Group Sparsity Constraints

Abstract

The invention discloses an unsupervised object recognition method combining multi-source feature learning and group sparsity constraints, which includes the following steps: Step 1, obtaining V types of views from an image set to be processed containing c categories, and forming them into a data set X =[x 1 , x 2 ,...,x n ]∈R d×n , where d represents the feature dimension of the data, and n represents the number of samples in the data set; step 2, extract the total scatter matrix S of the data set X t ; Step 3, building a KM clustering model based on linear discriminant analysis on the basis of step 2; Step 4, building a multi-source data joint clustering model based on group sparse constraints and feature selection on the basis of step 3; Step 5, Solve the objective function of the multi-source data joint clustering model obtained in step 4, and optimize it. This method can improve the accuracy of the clustering method, quickly locate the optimal feature subset, and effectively suppress the noise interference in the data set, and finally provide effective support for machine learning and computer vision related applications.

Description

technical field [0001] The invention belongs to the technical field of machine learning, in particular to an unsupervised object recognition method combining multi-source feature learning and group sparse constraints. Background technique [0002] Clustering technology aims to divide the object to be processed into multiple similar clusters, and then extract the abstract semantics of the data. It is a very widely used technology and has achieved great success in the field of image object recognition. However, the traditional clustering methods in the field of image object recognition still have the following defects: first, image data is generally composed of high-dimensional features, and these high-dimensional image data often contain noise features, directly operating on them will seriously affect the clustering effect; Secondly, these high-dimensional image data generally have a large number of redundant features, and processing such data requires expensive computing res...

AI Technical Summary

Problems solved by technology

However, the traditional clustering methods in the field of image object recognition still have the following defects: first, image data is generally composed of high-dimensional features, and these high-dimensional image data often contain noise features, directly operating on them will seriously affect the clustering effect; Secondly, these high-dimensional image data generally have a large number of redundant features, and processing such data requires expensive computing resources

However, these methods need to use dimensionality reduction methods to locate the feature subspace, and the obtained subspace is quite different from the original space, which is difficult to meet the needs of data semantic understanding in practical applications.

In addition, in order to deal with multi-view data, such methods need to determine the optimal subspace for each view separately. When the number of views increases, the complexity of the subspace search will rise sharply

Images