基于Directionlet变换的遥感图像降噪算法研究  

Research on Algorithm of Remote Sensing Image Despeckling Based on Directionlet Transform

作　　者：李彦[1] 张德祥[2] 桂树国[1] 

机构地区：[1]安徽职业技术学院,安徽合肥230011 [2]安徽大学电气工程与自动化学院,安徽合肥230601

出　　处：《佳木斯大学学报（自然科学版）》2016年第4期560-563,共4页Journal of Jiamusi University：Natural Science Edition

基　　金：国家自然科学基金(61272025)

摘　　要：提出一种有效的基于Directionlet变换的Bayesian最大后验估计实现遥感图像降噪算法。首先采用具有多尺度、多方向特点的Directionlet变换对含噪遥感图像进行分解,然后根据先验信息和噪声系数分布设计最大后验估计算法实现相干斑噪声的去除。实验结果表明,提出的算法与传统降噪算法相比不仅原始图像的边缘和纹理信息可保留,而且可获得更好的降噪视觉效果。An efficient despeckling algorithm for remote sensing images based on Directionlet transform using Bayesian the maximum a posteriori （MAP） estimator is proposed. Firstly, the noise remote sensing images are decomposed using Directionlet transform, which have multi - scale, multi - direction characteristics. Then, the maximum a posteriori estimation algorithm is designed to achieve the removal of speckle noise according to the prior information and the distribution of noise coefficient. Experimental results show that compared with traditional algorithm, the proposed algorithm can get better visual effect and the significant information of original image like textures and contour details is well maintained.

关 键 词：Directionlet变换 遥感图像 Bayesian最大后验估计 图像降噪 

分 类 号：TN957.54[电子电信—信号与信息处理]