基于邻域质心投票的分类影像道路中心线提取原理.DOC

中图法分类号：P407.8 文献标识码： 文章编号： 论文引用格式： 利用邻域质心投票分类影像 提取道路中心线 丁磊，，郭海涛，刘志青 信息工程大学，河南 郑州 450001 高分辨率影像道路二值了道路，且道路呈直接应用于生产与。邻域质心投票道路算法各方向的距离邻域多边形，进行质心投票来提取道路的中心线，同时估算道路宽度并判断出连通距离较长的方向数目以排除非道路区域的干扰经得到的中心线。图像及不同特征的结果，该算法与Zhang和Couloigner提出的进行了。该算法为80.6%79.0%，且计算效率，的小于参考算法的20%在稳定性及不同路宽的适应性多个方面。一种传统方法中提纯与提取功能从分类影像直接提取道路中心线，能够形状特征具备一定抗能力，适用于非道路区域的分类结果。关键词：；；；；；  Using Neighborhood Centroid Voting to Extract Road Centerline from Classified High Resolution Image DING Lei, YAO Hong, GUO Haitao, LIU Zhiqing Information Engineering University, Zhengzhou, 450001 Abstract: Objective: While applying image classification algorithms on high-resolution images to extract roads, non-road areas do exist in the binary result. Meanwhile, the achieved roads are planar, which can not be used directly for production and research purposes. In this case, a novel algorithm named neighborhood centroid voting is proposed to extract road centerlines. Method: Firstly, build a neighborhood polygon for each road pixel by detecting the connective distance in each direction. After that, vote for centroids of these polygons to extract road centerlines, at the same time estimating the road width and recording the number of those directions with a comparatively long connective distance to exclude none-road areas. Finally, applying morphological methods to obtain thinned centerlines. Result: A comparison is made between this algorithm and a reference method proposed by Zhang and Couloigner by means of experiments on a test image and two classified high resolution aerial images with different road distributions. Results suggest that the quality of this algorithm is 80.6% and 79.0%. Taking less than 20% the time of the reference method for dealing with actual images, this algorithm has a strong advantage in effectiveness. Additionally, this algorithm shows a higher stability and an adaptation for roads with different width. Conclusion: The proposed algorithm named neighborhood centroid voting is a centerli