摘要
提出了一种新的基于边缘特征的合成孔径雷达(SAR)图像与光学图像相匹配的算法,可用于无人飞行器的导航。首先针对SAR图像低信噪比的特定情况,提出了一种简单实用的线状特征提取方法。而后在提取边缘的基础上,采用改进的具有强抗干扰能力和容错能力的Hausdorff距离作为相似性准则进行图像匹配。在搜索策略上,采用遗传算法来加快搜索的速度。实验结果表明,这种算法匹配精度高,计算速度快。
A novel algorithm for matching synthetic aperture radar (SAR) image with optical imagebased on lineal feature using modified Hausdorff distance combined with genetic algorithm is proposed inthis paper. A new method to extract lineal feature from low SNR SAR image is presented firstly. Based onthe edge image from SAR and optical images, modified Hausdorff distance is adopted as a similarity measurebecause it is insensitive to noise. In searching strategy. genetic algorithm is used to achieve high computa-tion speed. Experiment results using real SAR and optical images demonstrate that the algorithm is robustand fast, and can achieve high matching accuracy.
出处
《雷达科学与技术》
2003年第4期242-245,254,共5页
Radar Science and Technology
