基于圆形区域拟合匹配的多图幅接边算法研究

Research of multi-map junction based on circular area fitting method

针对矢量型线要素图幅接边中常常出现的由于边界要素间距离计算量大导致程序效率低的问题,提出了一种基于距离匹配和趋势增强的圆形区域拟合匹配接边算法。该算法利用边界线缓冲区来获取候选匹配要素集,然后通过为候选匹配集中要素端点生成圆形缓冲区来筛选相邻图幅中与该要素满足距离条件的要素,最后通过为距离匹配要素生成拟合直线来获取要素在边界附近的趋势走向,从而获得与某要素趋势最为相近的边界要素;该算法还利用图幅结合表实现了多图幅自动接边。实验证明,所提出的算法较之传统的图幅边界处同名实体几何匹配算法,有效地减少了相邻图幅间要素距离的计算量,提高了实体匹配的效率与准确率,是切实可行的。

In the work of vector linear feature map merging, the problem of poor efficiency because of the large amount of dis- tance calculation between the boundary elements often exists. This paper proposed a circular area fitting matching algorithm, which based on the distancematching and the edge trend enhanced method. Firstly ,this algorithm used boundary line buffer to obtain the candidate matching feature sets. Then, it selected features which met the conditions of distance on the adjoining sheet by generating a circular buffer for endpoints in the candidate matching feature sets. Finally,it obtained the tendency of the dis- tance matching features by generating fitting straight lines, by comparing slopes of the fitting straight lines to obtain the feature which had the most similar trend. It also achieved multi-map automatic junction by map sheet combined table. Experiments show that compares with the traditional geometric matching methods of the identical entity, it reduces the computation of the distance between the features and improves the efficiency and accuracy of elements matching effectively. The feasibility of the algorithm is validated by practical application.