基于特征模板匹配识别地图中特殊图标的方法

Method based on feature template matching to identify special icons in map

将栅格地图作为图源来生成矢量电子地图的关键一步,就是首先识别交通地图上的特殊图标。交通地图中有许多像医院、电信、政府等特出图标,如何识别这些图标,实现在矢量地图中对以上信息的自动添加,一直是业内研究的课题之一。基于特征模板匹配法,从颜色特征着手,利用最大最小距离算法和颜色一致性聚类分析,实现对栅格地图中特殊图标的快速、准确识别。以医院图标为例,介绍识别过程和软件实现,最后给出仿真结果。

Raster map as figure source to generate the vector electronic map key step is the first to recognize the special icon on the traffic map. There are many outstanding symbols such as hospitals, telecommunications, government on the traffic map. These has been the subject of study how to identify these symbols and automatically add the above information in the vector map. Based on matching method of the feature template from the color features to examine, us- ing maximum and minimum distance algorithm and color consistency of cluster analysis, the special icons are quickly and accurately identified on the grid map. In this paper, the hospital icon is as an example, the recognition process and soft- ware implementation are introduced, and the final simulation results are gotten.