SAR image ship target detection method

Abstract

The invention discloses an SAR image ship target detection method, and implementation steps thereof include: determining an SAR ship image to be input and carrying out clutter suppression; then usingan improved SLIC superpixel generation algorithm to segment the SAR ship image into super-pixel blocks, and then calculating self-information values of each superpixel block and setting a threshold value T1 to picking out candidate superpixel blocks; calculating expanding neighborhood weighted information entropy in four directions of candidate superpixel patches, and setting an expanding neighborhood weighted information entropy growth rate T2 to remove false alarm candidate superpixel patches; and performing Harris corner detection on a detection result, and setting the number T3 of cornersto further filter out the false alarm patches, so as to obtain a final SAR image ship target detection result. The SAR image ship target detection method provided by the invention fully utilizes the combination of superpixel segmentation, information theory detection and Harris corner detection to realize SAR image ship target detection, and the obtained detection result shows that the method provided by the invention has the advantages of high correct detection rate, low false alarm rate and omission rate, and low detection distortion rate.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to a SAR image ship target detection method and system based on the combination of information theory and Harris corner detection. Background technique [0002] Synthetic Aperture Radar (SAR) imaging is not limited by weather, light and other conditions, and can conduct all-weather and all-weather reconnaissance of interested targets. In 1978, for the first time, the United States clearly obtained ship images and other information from the SAR images of the Seasat-1 satellite. It is very important to carry out research on surface ship target detection for maritime monitoring management and timely acquisition of military intelligence information. [0003] There are many methods for ship target detection, such as segmentation method, fractal theory, wavelet decomposition, template, likelihood ratio, multi-polarization data, constant false alarm rate (Constant False Alarm Rat...

AI Technical Summary

Problems solved by technology

And combined with superpixel segmentation and information theory detection, the original shape of SAR ship target can be well preserved. If it is replaced by target detection based on CFAR combined with morphological processing, the ship image detection results in the detection results will often be distorted. , at this time, the corner detection result of the detection result is not the corner detection result of the ship target in the original SAR image, which will affect the detection accuracy

Images