摘要： 针对多尺度面状要素的一对多匹配问题，提出基于正切空间的多尺度面实体形状相似性匹配方法。通过源多边形外扩缓冲区搜索候选匹配集，对候选匹配集中的面要素进行组合，组合生成的多边形与源多边形进行形状匹配，从而获得最佳匹配集。实验结果表明，该方法计算过程简单，容易实现且匹配效率较高。

关键词: 空间相似性, 矢量匹配, 正切空间, 多尺度

Abstract: Aiming at the problem of one-to-many matching for multi-scale facet element, this paper presents multi-scale area entity shape matching based on tangent space method. It searches the candidate matching set by expanding the buffer zone outside the source polygon, combines the area patterns in the candidate matching assemble, matches the new polygon shape with the original polygon shape after combining the area patterns, and gets the best matching set. Experimental results show that computational process of the method is simple and easy to implement, and it has better matching efficiency.

Key words: space similarity, vector matching, tangent space, multi-scale

中图分类号: 

• TP393