SAR image ship target detection method

Abstract

The invention discloses an SAR image ship target detection method, which relates to the technical field of SAR image target detection, and comprises the following steps of: (1) selecting a threshold value according to the occurrence frequency of a pixel gray value to screen clutter pixels, calculating a kernel density value of each pixel point in an SAR image based on a selected sliding window, and converting into spatial values corresponding to the pixel points through normalization; (2) performing resolution reduction processing on the image through a sampling theorem; and (3) realizing self-adaptive CFAR detection based on G0 distribution in the resolution-reduced image subjected to pixel screening. According to the method, the image is subjected to resolution reduction processing before the joint image is detected, so that the SAR image is suitable for existing CFAR image processing, the detection efficiency can be improved by reducing the resolution, the false alarm rate can be kept constant and reduced, and meanwhile, the accuracy of a detection numerical value can be improved.

Description

technical field [0001] The invention relates to the technical field of SAR image target detection, in particular to a SAR image ship target detection method. Background technique [0002] The launch of the ocean satellite SeaSat-1 in the 1970s marked the world's entry into the era of modern space remote sensing, and ocean observations entered the era of stereoscopic observations. Ocean satellites and satellite remote sensing ocean applications have become the main means of modern ocean surveillance. In marine surveillance applications, SAR, especially spaceborne SAR, has become a useful supplement or even an alternative to shore-based radar and optical sensors. In recent years, as an important part of marine surveillance applications, automatic detection of ship targets based on SAR images has increasingly become a research hotspot. The CFAR and two-parameter CFAR ship target detection methods based on the background clutter statistical model are both pixel-level detections...

AI Technical Summary

Problems solved by technology

However, most of the traditional CFAR detection methods are aimed at low-resolution SAR images. With the improvement of SAR imaging capability and image resolution, the traditional CFAR detection methods can no longer meet the requirements of automatic detection of ship targets.

Images