In the process of SAR (Synthetic Aperture Radar) ATR (Automatic Target Recognition), man-made target detection is very important. In this paper, analyzing the characteristics of SAR images, a new method is proposed to detect man-made targets in SAR images. Firstly, the image is smoothed with the adaptive wavelet filter; the multiple speckles are removed while the edge property of SAR images is sustained. It made the detection easier. Then, the images is segmented with the 2-D C-partition fussy entropy method based on adaptive genetic algorithm. Test results show that the algorithm can detect man-made targets effectively.
Lee Jong-Sen. Speckle analysis and smoothing of synthetic aperture radar images[J].Computer Graphics and Image Processing.1981, 17(1):24-32[2]Donoho D L. De-noising by soft-thresholding[J].IEEE Trans. on Information Theory.1995, 41(3):613-627[3]Arsenault H H, April G. Properties of speckle integrated with a finite aperture and logarithmically transformed. J. Opt, soc, 1976: 1160-1163.[4]Touzi R, Lopes A, Bousquet P. A statistical and geometrical edge detector for SAR images[J].IEEE Trans. on Geoscience and Remote Sensing.1988, 26(7):764-773[5]金立左,袁晓辉,赵一凡,夏良正. 二维模糊划分最大熵图像分割算法[J].电子与信息学报.2002, 24(8):1040-1048浏览[6]王小平,曹立明. 遗传算法-理论、应用与软件实现. 西安:西安交通大学出版社,2002: 73-74.