Contourlet域的多极化SAR图像自适应融合方法  

Adaptive fusion method for multi-polarization SAR images in Contourlet domain

作　　者：孟庆文[1] 周焰[1] 徐颖[1] 

机构地区：[1]空军预警学院,武汉430019

出　　处：《空军预警学院学报》2014年第4期265-268,275,共5页Journal of Air Force Early Warning Academy

摘　　要：针对不同极化SAR图像的融合,提出了一种基于Contourlet变换的自适应窗口图像融合方法.首先对原始图像进行Contourlet变换,将图像分解为一个低频子带和多个不同方向的高频子带.对分解后的低频子带进行邻域能量加权融合,对各方向高频子带采用依据SAR图像斑点噪声特征的变化自适应地调整融合窗口的方法.通过对两极化SAR图像进行融合实验并与小波变换等融合结果比较,表明本文方法融合后的图像在视觉特性以及客观评价统计因子上取得了更好的效果,融合后的图像提供了比原始图像更丰富的信息.In consideration of the fusion of multi-polarization SAR images, this paper proposes a method of adaptive window image fusion based on Contourlet transform. In this scheme, the original image is Contourlet transformed at first, the image is decomposed into a low frequency (LF) sub-band and a few high frequency (HF) sub-bands on various directions, and then the decomposed LF sub-bands are fused in the weighting of neighborhood energy, next, the adaptive adjusting fusion window method could be used for HF sub-bands on each direction, which id adaptive to the change of speckle noise characteristics of SAR image. By doing the fusion experiments on bi-polarization SAR images and comparing with Wavelet transform and other fusion results, it is shown that the images fused by this proposed method could achieve the better effect in terms of visual characteristics and the statistical factor of objective evaluation, and the fused images could provide more abundant information than that of the originals.

关 键 词：SAR图像融合 CONTOURLET变换 自适应窗口 多极化SAR图像 

分 类 号：TN958[电子电信—信号与信息处理] TP751.1[电子电信—信息与通信工程]