基于水平集的SAR图像与光学图像的配准  

The Registration between SAR and Optical Image Based on Level Set

作　　者：李雨谦[1] 皮亦鸣[1] 王金峰[1] 

机构地区：[1]电子科技大学电子工程学院,四川成都610054

出　　处：《测绘学报》2010年第3期276-282,共7页Acta Geodaetica et Cartographica Sinica

基　　金：国家自然科学基金(60772143);国家863计划(2007AA12Z159)

摘　　要：为解决常用图像配准算法中特征提取易受噪声影响及配准过程受特征提取精度影响的问题,提出一种新的图像配准算法,即在水平集框架下,通过映射函数将SAR图像特征和光学图像特征结合,构造能量泛函模型,并采用水平集方法求解曲线演化方程,进而实现SAR图像和光学图像的配准。通过两个场景的配准实验表明,该算法不仅对高信噪比SAR图像和光学图像的配准有效,且对低信噪比的配准优势更加明显。A novel algorithm for the image registration between synthetic aperture radar（SAR）image and optical image is proposed.With this approach,the accuracy of registration won＇t be affected by speckle noises or the result of feature extraction.Under the level set framework,an energy function is defined with the matching function between two images,and the function can be solved by curve evolution.The efficiency of the algorithm is verified by the real SAR images and optical images of two different scenes.Experimental results reveal that the registrations between multi-sensors are efficient in both high and low SNR images.

关 键 词：图像配准 水平集 SAR 光学图像 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]