Image classification method based on subspace projection and dictionary learning

Abstract

The invention discloses an image classification method based on subspace projection and dictionary learning. Firstly, a discrimination dictionary is initialized through a training set sample with a label, and then a class label of a test sample is predicted by using the discrimination dictionary. A test set sample with high reliability is selected and a low dimensional subspace is learned from thetest set sample with false tags and the discriminant dictionary is updated in this low dimensional space. The updated discriminant dictionary is used to reclassify the test set samples, the pseudo-label obtained by this iteration is compared with the pseudo-label obtained by the previous iteration, the samples with the same attribute obtained by the two iterations are called stable samples. If the number of stable samples exceeds eighty percent of the number of samples in the test set after one iteration, the pseudo-label obtained by this iteration is output as the classification result afterthe iteration is over. Compared with the prior art adaptive image classification method, the algorithm of the invention can obtain higher classification accuracy.

Description

technical field [0001] The invention relates to the field of domain-adaptive image classification in pattern recognition, in particular to the field of domain-adaptive image classification based on subspace projection and dictionary learning. Background technique [0002] In traditional pattern recognition and machine learning, the algorithm needs to perform well under certain assumptions. The most common one is that the distribution of the training set samples in the source domain and the test set samples in the target domain must be consistent. However, in practical applications, it is often difficult to satisfy this condition. For example, in the field of image classification, different types of image sensors, different shooting angles, different lighting conditions, etc., will cause a large distribution difference between the training set and the test set samples. The difference in distribution will cause the trained classifier to perform poorly when actually processing...

AI Technical Summary

Problems solved by technology

[0004] The technical problem to be solved by the present invention aims at the defects involved in the background technology, and provides an image classification method based on subspace projection and dictionary learning, which solves the problem of training set and source domain without retraining the classifier. The problem of poor performance of the classifier caused by the excessive distribution of test set samples in the target domain

Images