基于Contourlet能量金字塔和改进Hausdorff距离的SAR匹配图像算法

摘要

针对SAR图像匹配系统的高精度、高效率要求,对SAR图像匹配方法进行了研究.将Contourlet分解的金字塔结构应用于SAR图像匹配并对其进行了改进,形成了基于Contourlet变换的能量金字塔算法,同时针对相似性度量方法,重点研究了鲁棒Hausdorff距离(RHD)出格点阈值的选取方法,对Hausdorff距离进行了改进,进而提出了基于Contourlet能量金字塔和改进的Hausdorff距离的SAR图像匹配算法.经过实验对比,证明了这种新算法不仅鲁棒性较好而且匹配精度高,能够达到SAR图像匹配系统的要求.%SAR image matching was studied to improve the matching precision and the matching efficient of SAR image matching systems.The pyramid structure based on the Contourlet decomposition was applied to SAR image matching and the energy pyramid image matching algorithm based on Contourlet transform was presented.In addition, aiming at the similarity measurement methods, the robust hausdorff distance ( RHD) was studied emphatically, and a RHD outlier point threshold selection method was proposed and the Hausdorff distance (HD) was improved.And then, a new SAR image matching algorithm based on the Contourlet energy pyramid and the improved Hausdorff distance was put forward.The experiments and comparisons validate the better robustness and high matching precision of the new algorithm, so it can meet the requirements of SAR image matching systems.