Self-adaptive moving target detection method capable of combining polarized classification and power grouping

Abstract

The invention relates to a self-adaptive moving target detection method capable of combining polarized classification and power grouping, mainly aiming at solving the problems that in the non-uniform scene, the clutter covariance matrix is low in estimation accuracy and poor in moving target detection performance. The method comprises the steps of: 1, synthetic aperture radar (SAR) image polarized classification; 2, initialized-power grouping; 3, group sample number check; 4, self-adaptive clutter reduction; and 5, constant false alarm rate detection. By adopting the polarized classification method to guide the screening of samples, the method solves the problem that the clutter characteristic difference is not considered in the existing sample screening method, thus having the advantages that the clutter covariance matrix is high in structure estimation accuracy and good in moving target detection performance. By adopting the power grouping method to guide the screening of samples, the method solves the problem that the existing sample power selection is overhigh in power estimation, thus having the advantages that clutter covariance matrix is high in power estimation accuracy and good in low-speed moving target detection performance.

Description

technical field [0001] The invention belongs to the technical field of radar signal processing, and further relates to a sample screening method combined with synthetic aperture radar SAR image polarization classification and power grouping guidance in the technical field of moving target detection. The invention can be applied to the ground of synthetic aperture radar on moving platforms moving target detection. In the absence of prior knowledge, the present invention can select samples similar to the clutter characteristics of the unit samples to be detected through polarization classification and power grouping, thereby improving the estimation accuracy of clutter covariance matrix and the detection performance of moving targets. Background technique [0002] Synthetic Aperture Radar Ground Moving Target Detection Technology (SAR-GMTI) can detect moving targets in the observation area by using multi-channel SAR image data combined with adaptive moving target detection met...

AI Technical Summary

Problems solved by technology

This method uses the SAR image polarization information to classify each pixel unit and the corresponding samples, and performs power grouping on the basis of the classification results, which improves the problem of poor estimation accuracy of the clutter covariance matrix caused by sample differences, and improves the self-efficacy. Adaptive Processing for Moving Object Detection Performance

Images