论文部分内容阅读
提出了一种新的完整地利用高分辨率光学图像和SAR图像进行城市矩形建筑物检测和高度估计的方法.该方法充分利用了两种数据的互补性,由三个顺序执行的步骤组成:首先对两种数据分别进行建筑物检测,然后基于一种整合的准则对两种图像进行配准,最后利用一种“假设生成-仿真-匹配”的方法对建筑物高度进行估计.该方法对部分遮掩的建筑物也能取得好的效果.对两个不同场景的测试验证了该方法的有效性.
A new method for the detection and height estimation of urban rectangular buildings using a complete set of high resolution optical images and SAR images is proposed.The method makes full use of the complementarity of two kinds of data and consists of three sequential steps: First, the two kinds of data are respectively tested for buildings, then the two images are registered based on an integrated criterion, and finally the building height is estimated by using a method of “hypothetical generation-simulation-match” The method can also achieve good results for partially concealed buildings.The testing of two different scenarios verifies the effectiveness of the proposed method.