SAR target identification method based on image bit hierarchical interpretation

Abstract

The invention belongs to the field of SAR (Synthetic Aperture Radar) image interpretation and target identification, and particularly relates to an SAR target identification method based on image bithierarchical interpretation. An original SAR image can inhibit SAR image speckles through bit hierarchical processing. Local feature information of the SAR image can be kept. Identification features are extracted from the SAR image subjected to hierarchical interpretation through a feature extraction method, and the identification precision can be effectively improved; the method is especially suitable for high-pitch-angle SAR image recognition. According to the invention, bit hierarchical interpretation is used to complete preprocessing of an original SAR image. The inter-class difference isimproved. The situation that the speckle noise of the image is large after the pitch angle of the SAR image is increased can be effectively solved, and therefore SAR image target identification with high accuracy can be achieved for a high-pitch-angle SAR image data set.

Description

technical field [0001] The invention belongs to the field of synthetic aperture radar (Synthetic Aperture Radar, SAR) image interpretation and target recognition, in particular to a SAR target recognition method based on image bit layered interpretation. Background technique [0002] With the continuous development of SAR imaging technology, high-resolution SAR images have been widely used in the field of remote sensing; SAR target recognition is one of the research hotspots of synthetic aperture radar. Feature extraction and classifier design are two crucial operations in SAR target recognition. However, due to the unique SAR coherent imaging mechanism, SAR images contain complex information such as target, target back, clutter, and background. There are significant challenges in robust image feature extraction. [0003] In order to suppress the influence of SAR image coherent speckle on SAR target recognition, early scholars have proposed image filtering algorithms such a...

AI Technical Summary

Problems solved by technology

Feature extraction and classifier design are crucial two-step operations in SAR target recognition, but due to the unique SAR coherent imaging mechanism, SAR images will contain complex information such as targets, target backgrounds, clutter, background, etc. Significant challenges in image robust feature extraction

[0003] In order to suppress the impact of SAR image coherent speckle on SAR target recognition, early scholars have proposed image filtering algorithms such as Lee, Kuan, Frost filters, etc.; but these algorithms have parameter estimation problems, and these filtering algorithms may lose the SAR image. Local feature information; and the local information of the target is very useful for SAR images, especially for SAR image recognition collected under conditions of large elevation angles

Images