SAR Target Recognition Method Based on Non-negative Least Squares Sparse Representation

Abstract

The invention provides a SAR target recognition method based on non-negative least squares sparse representation, which is based on the spectral features of SAR images as recognition features, by projecting test samples onto the training set, and adding non-negative Negative constraints avoid interference to radar target recognition due to the fact that the sparse coefficients in the sparse representation are positive or negative, and the mathematical description of the sparse representation is unrealistic. At the same time, the sparse solution can more effectively reflect the low dimensional structure, through the sparse reconstruction process to determine the category of the test sample, realize the recognition of radar targets, thereby improving the recognition rate, while avoiding the azimuth estimation of SAR image targets and factors such as defocus or signal-to-noise ratio. The interference caused by the radar has good noise robustness, which can effectively improve the accuracy of radar target recognition.

Description

technical field [0001] The invention relates to the technical field of radar target recognition, in particular to a SAR target recognition method based on non-negative least square sparse representation. Background technique [0002] Synthetic Aperture Radar (SAR) technology is a pulse radar technology that uses mobile radar mounted on satellites or aircraft to obtain radar target images in high-precision geographic areas. Synthetic Aperture Radar Auto Targets Recognition (SAR-ATR) has important application value in many geographic information analysis technology fields. [0003] The recognition performance of automatic radar target recognition is mainly determined by feature extraction and recognition algorithms. In terms of feature extraction, better feature extraction can not only reduce the dimensionality of data recognition, but also retain as much effective information as possible for recognition. Due to some unique characteristics of SAR images, such as specular ref...

AI Technical Summary

Problems solved by technology

[0005] In view of the above-mentioned problems existing in the prior art, in order to solve the problem that the SAR image target recognition in the prior art needs to estimate the target azimuth angle and the recognition accuracy is limited, the present invention provides a SAR based on non-negative least squares sparse representation Target recognition method, the radar target recognition method obtains sparse vectors by projecting test samples onto the training set, and then determines the category of test samples through the sparse reconstruction process to realize the recognition of radar targets, and in the process of sparse projection By adding non-negative constraints, the results of projection reconstruction are more in line with the actual situation, and the obtained sparse solution will be more accurate and sparse, so that the recognition of radar targets does not need to rely on the target azimuth estimation of SAR images At the same time, it can avoid the interference caused by factors such as defocus or signal-to-noise ratio on target recognition, and improve the accuracy of SAR target recognition

Images