提出了一种基于道路中心线的闭环反馈改善交通地图图像中道路识别与提取的方法,以达到识别与提取完整道路网络的目的.依据于微观识别与宏观排除的逼近思想,实现了道路图层的初始聚类与输出反馈控制下的道路图层完全提取.针对标准城市交通地图,确定了道路与区域阈值条件.通过对噪声的再聚类过程实现道路与区域的二值完全聚类.利用交通道路中心线特征构成道路图层反馈再聚类策略.反馈再聚类策略能够保证道路图层优化过程的收敛性质.试验结果表明了该方法具有较高的准确性、全自动化和通用性.%This paper introduces a feedback closed loop method to recognition automatically and extraction roads from color city maps, in order to recognitize and extract the road network fully. Based on approach thought of microscopic identification and macroscopic elimination, the system realized first the road layer initial recognition and last road map level completely to withdraw via some output layers reclustering and feedback control. In view of the standard municipal transportation map, the path and the region valve value condition has determined. Through re-clustering noise the whole map can be completely divided into two parts of road and non-road(region). The feedback re-clustering strategies of recognition for transportation road networks are based on the core characteristic of obtained road constitution. The convergence of the feedback re-clustering of road layer guarantees the map level optimization. The experiments prove the method is an efficient method with advantages of precision, automation and broad use.
展开▼
