Radar target imaging and recognition method based on support set constraints

Abstract

The invention discloses a radar target imaging and recognition method based on support set constraints. The method comprises the steps that: support set constraint conditions are built according to prior information on the basis of a wavefront reconstruction algorithm of radar target imaging, and a compressed sensing radar target imaging model based on the support set constraints is obtained; andthen a quadratic support set updating method is adopted to improve an orthogonal matching pursuit reconstruction algorithm, the compressed sensing radar target imaging model based on the support set constraints is reconstructed to achieve the purpose of focusing imaging; and finally target recognition is realized by adopting a support vector machine (SVM) learning method based on a high-resolutionimaging result. Due to the fact that the support set constraint conditions are added, better imaging quality can be obtained under a low signal-to-noise ratio compared imaging quality obtained with arandom support set. And on the basis, the marked target high-quality imaging result is used as a training sample to train an SVM learner, and finally an unknown target imaging result is sent to the SVM learner to realize effective target identification.

Description

Technical field[0001]The invention belongs to the technical field of signal and information processing, and specifically relates to a radar target imaging and recognition method based on support set constraints.Background technique[0002]With the development of information technology and signal processing algorithms, radar target imaging and recognition have become more and more widely used in areas such as intelligence reconnaissance, sea surface monitoring, and security inspection. At the same time, people have higher and higher requirements for target imaging quality and target recognition rate. However, in actual engineering, when performing radar target imaging and target recognition, there are a wide range of noise sources, which will have a greater impact on target imaging and recognition results. In order to effectively improve the imaging quality of radar targets and the noise immunity of target recognition, scholars have carried out many researches on radar target imaging t...

AI Technical Summary

Problems solved by technology

[0003] However, most of the existing research on compressed sensing radar imaging algorithms is limited to the establishment of imaging models and the optimization of signal reconstruction algorithms, without considering the limitation of the support set during signal reconstruction, which makes it difficult for the algorithm to meet high Quality imaging requirements, it is difficult to provide effective feature information for subsequent target recognition

Images