Abstract: Aiming at extracting building height from high resolution SAR image, a retrieval method based on an highlight model matching technique is proposed. Highlight feature model is established by analyzing building imaging characteristics, and the matching function between the new model and the SAR image is defined. Then the matching function is optimized by multi-population genetic algorithm and optimal height is searched. Based on the experiment of simulation and real SAR image, it is proved that the method can be used in the building height retrieval with good retrieval accuracy.

Key words: high resolution Synthetic Aperture Radar（SAR） image, highlight feature model, matching function, building height retrieval

摘要： 针对高分辨率SAR图像中的建筑物高度提取问题，提出了一种基于高亮模型匹配的建筑物高度反演方法。通过对建筑物的成像特征进行分析构建出高亮特征模型，建立模型与SAR图像之间的匹配度函数，运用多种群遗传算法对匹配度函数进行优化搜索出最优的高度参数。基于模拟和实测SAR图像的实验结果表明该算法可以用于SAR图像建筑物高度反演，并具有较高的反演精度。

关键词: 高分辨率合成孔径雷达（SAR）图像, 高亮特征模型, 匹配度函数, 建筑物高度反演