论文部分内容阅读
针对无矢量参考数据自发地理信息道路精度难以评价的问题,提出了一种基于影像匹配的道路精度评价方法。首先提取影像与自发地理信息中的道路交叉口作为控制点,并将对应的交叉口进行匹配,以同名控制点的均方根误差作为自发地理信息的道路精度;然后以两组控制点分别构建Delaunay三角网,利用两组控制点的对应关系对每个三角网进行仿射变换,从而实现对自发地理信息道路的几何纠正,以提高其精度。最后以郑州市的Open Street Map道路数据进行试验,结果表明本文算法能够有效提高自发地理信息的道路精度。
Aiming at the problem that it is difficult to evaluate the road accuracy of spontaneous geographical information without vector reference data, a road evaluation method based on image matching is proposed. Firstly, the road intersections in images and spontaneous geographic information are extracted as control points, and the corresponding intersections are matched. The root mean square error of control points with the same name is used as the road precision of autonomous geographical information. Then, two sets of control points are respectively constructed Delaunay triangulation network, the triangulation network is affine transformed by using the correspondence between two sets of control points, so as to realize the geometrical correction of spontaneous geographical information road to improve its accuracy. Finally, the test of Open Street Map in Zhengzhou City is carried out. The results show that this algorithm can effectively improve the road accuracy of autonomous geographic information.