【摘要】 针对现有利用阴影长度法提取建筑物高度时存在的阴影间相互遮挡问题,提出了一种基于建筑物侧面轮廓线进行建筑物高度估算的新方法。首先,利用RPC模型计算建筑物像点位移的方向与卫星成像角度,再将遥感影像进行旋转,使建筑物像点位移沿水平方向;然后,利用Canny算法进行轮廓检测,并构建一定长度的矩形形态学结构元素,对轮廓图像进行形态学开运算,以提取侧面轮廓线,再利用Hough变换与建筑物角点约束,对所提取的轮廓线进一步筛选;最后,根据卫星侧视成像时建筑物高度与像点位移的几何关系进行建筑物的高度估算。利用实际的高分辨率卫星影像对本文方法进行了验证,并与阴影法估算建筑物高度进行了对比。试验结果证明,利用建筑物侧面轮廓线进行建筑物高度估算平均误差可以达到0.7 m,且实际精度优于使用阴影法进行建筑物高度估算。更多还原

【Abstract】 In view of the mutual occlusion between shadows when we use shadows to calculate the height of a building,in this paper,we propose a new method for building height estimation based on building profile. First,we use RPC model to calculate the direction of image displacement and satellite imaging angle,then rotate the remote sensing image to make the building image displacement along the horizontal direction. Then we use the Canny algorithm for contour detection and construct rectangular morphological structural elements to perform morphological opening operations on contour images to extract side contours. Subsequently Hough transform and building corner are used to further screen the extracted contours. Finally,the building height is estimated according to the geometric relationship between the building height and the displacement of the image point. We use high-resolution satellite images to verify the method and compare it with the shadow method to estimate the building height. The experimental results indicate that the average error of building height using the profile line can reach 0.7 m,and the actual accuracy is better than using the shadow method.更多还原