基于Contourlet尺度相关的SAR图像融合

针对不同波段全极化SAR图像的融合,提出了一种基于Contourlet变换域进行图像融合的方法.利用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;对方向高频子带依据尺度间相关性定义一个边缘信息量测指标,选择量测指标大的系数作为融合系数.最后对两波段全极化SAR图像进行了融合实验.实验结果表明：与小波变换等融合方法相比,利用所提出方法融合后的图像能够明显抑制SAR斑点噪声影响,在视觉特性以及评价指标上效果更好.

Fusion of SAR Images Based on Contourlet Scale Correlation

For the fusion of polarimetric SAR image in various bands,this paper presents an image fusion method,based on the Contourlet transform domain.The proposed scheme,first,decomposes the image into a low frequency sub-band and multiple high frequency sub-bands in various orientations,next,makes the fusion of the decomposed low frequency sub-band with neighborhood energy weight,and then defines an edge information measurement index for the directional HF sub-band in terms of scale correlation,selects a coefficient whose measurement index is greater as the fusion coefficient and finally performs a fusion experiment on two-band polarimetric SAR images.Experimental results show that the fusion image by using the proposed scheme can greatly reduce the noise impact of SAR speckles,and have much better effect on visual characteristics and the evaluation indexes.