摘要
基于建筑物的OAR模型,建立一种在单视角高分辨率航空影像中自动提取高层建筑物的策略.以建筑物垂直边缘为关键属性,结合窗户纹理和屋顶轮廓直线段属性及它们之间的关系,综合推理出建筑物的位置.实验结果表明,该方法能有效实现高层建筑物的自动提取.
The extraction of building shapes is difficult with traditional modeling and matching algorithms because of their diversities and complexities. In this paper, a strategy for automatic extraction of high-rise building in monocular high-resolution aerial images is developed based on the OAR model of buildings. The vertical edges(as the key attribute), the window textures,line segments of roof contour and their relationships are used to locate the buildings. Experiment results have shown that this proposed model can extract the high-rise buildings automatically and effectively.
出处
《深圳大学学报(理工版)》
EI
CAS
北大核心
2005年第1期1-6,共6页
Journal of Shenzhen University(Science and Engineering)
基金
国家自然科学基金资助项目(60172066)
国家863高技术研究发展计划资助项目(2001AA136070)