一种改进的基于各向异性微分的SAR图像增强方法  

An improved enhancement approach based on anisotropy differential used for SAR image

作　　者：张齐[1] 曹琦[1] 毕笃彦[1] 王晓薇[2] 

机构地区：[1]空军工程大学工程学院,陕西西安710038 [2]空军工程大学电讯工程学院,陕西西安710077

出　　处：《光电子．激光》2010年第4期614-617,共4页Journal of Optoelectronics·Laser

基　　金：国家"863"计划资助项目(2007AA701121)

摘　　要：合成孔径雷达(SAR)图像经过相干斑抑制后,图像中目标边缘会出现模糊。针对这一问题,提出了一种改进的基于各向异性微分的SAR图像增强方法。该方法在梯度方向和与梯度垂直方向构造了两个扩散系数分布函数。在非目标边缘区域,各方向扩散程度不同,对相干斑噪声具有较好的抑制效果;在目标边缘区域,梯度方向不做平滑,以保护和增强边缘,而在垂直于梯度方向有较大的平滑,突出边缘轮廓。实验结果表明,该方法具有很好的相干斑抑制能力和边缘增强能力。Synthetic aperture radar（SAR） image after speckle suppression, the edge of the target image will appear fuzzy. In this paper,an improved enhancement approach based on anisotropy differential used for SAR image is proposed, two distribution functions are structured with gradient diffusion coefficient in gradient direction and vertical one in this paper. In nowtarget region,the extent of the spread is equal to all directions,and the speckle noise has a better inhibitory effect;in the target region,in order to protect and enhance edge,the edges in gradient direction are does not smoothed, and while in vertical gradient one,the edges have a greater smoothing highlight to perpendicular. Experimental results show that the method has both very good speckle suppression edge and enhancement capacity.

关 键 词：合成孔径雷达(SAR)图像 相干斑抑制 各向异性微分 边缘增强 图像增强 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]