Infrared target robustness recognition method in combination with multi-feature dimensionality reduction and transfer learning

Abstract

The invention discloses an infrared target robustness recognition method in combination with multi-feature dimensionality reduction and transfer learning. The method comprises: firstly providing, in view of the incomprehensive information coverage of a traditional infrared human target feature extraction method extracting a single feature, heterogeneous features for extracting different types of targets, including the shape feature and the brightness feature of the target image, so as to fully excavate the features of an infrared human target; secondly, using a principal component analysis method to reduce the dimensionality of the fused heterogeneous feature; and finally, designing an effective infrared human target classifier based on transfer learning to maximize the generalization performance and target recognition accuracy. Compared with the traditional infrared human target recognition method, the method provided by the invention improves the infrared human target recognition performance in a complex background.

Description

technical field [0001] The invention relates to a human body target recognition technology in an infrared scene, which effectively recognizes a human body target in an infrared scene, and belongs to the technical field of infrared image processing and pattern recognition. Background technique [0002] Infrared human target recognition is an important research branch in the field of infrared image processing and pattern recognition. Because in the infrared image, the human target is disturbed by imaging noise, the contrast is generally low; secondly, the background of the target is more complex, and the target is easily disturbed by the background; finally, the posture of the human target is changeable and easily blocked by other objects , therefore, the effective and robust recognition of infrared human targets is still a challenging subject, and its in-depth study has important theoretical significance and practical value. [0003] In order to effectively identify human ta...

AI Technical Summary

Problems solved by technology

First of all, in view of the problem that the traditional infrared human body target feature extraction method does not cover comprehensive information when extracting a single feature, from two different angles of shape and brightness, it is proposed to extract different types of heterogeneous features of the target, so as to fully excavate the infrared human body. characteristics of the target

Finally, in view of the lack of labeled infrared human target sample data, the poor generalization performance caused by the deviation of the distribution and semantics of training samples and test samples, an effective infrared human target classifier based on transfer learning is proposed. , which maximizes generalization performance and object recognition accuracy

Images