摘要
随着自动泊车、机器人服务等智能化行业的发展,以及室内服务需求的增加,传统手动方法与三角剖分法构建的路网模型难以满足高新技术作业的要求。因此,提出一种基于栅格的室内平面路网自动构建方法,该方法首先基于语义信息提取构建室内平面地图的必要要素,并将其转化为栅格;然后将栅格划分为可通行区域与不可通行区域,再通过深度优先搜索算法将室外区域障碍化;最后利用细化算法提取可通行区域的中心线,从而构建室内平面路网模型。通过对不同方法构建的路网进行对比分析,结果表明,该方法在完整度、相似度和构建时间成本上表现更佳。
With the development of intelligent industries such as automatic parking and robot services,as well as the increasing demand for indoor services,the road network models constructed by traditional manual methods and triangulation methods are no longer able to meet the requirements of high-tech operations.Therefore,we proposed an automatic construction method for indoor plane road network based on raster.We extracted the necessary elements for constructing indoor flat maps based on semantic information,and converted them into raster at first.Then,we divided the raster into passable and non-passable areas,and used depth first search algorithm to block the outdoor area.Finally,we used a refinement algorithm to extract the centerline of passable area,thereby constructed an indoor plane road network model.By comparing and analyzing road networks constructed by different methods,we found that this method performed better in terms of completeness,similarity and construction time cost.
作者
毛宏宇
效朝阳
蒋友龙
田仁涛
陈团
MAO Hongyu;XIAO Chaoyang;JIANG Youlong;TIAN Rentao;CHEN Tuan(The 8th Geological Brigade of Sichuan,Xichang 615000,China;School of Smart City,Chongqing Jiaotong University,Chongqing 400074,China)
出处
《地理空间信息》
2025年第6期47-50,共4页
Geospatial Information
关键词
栅格地图
室内路网
细化算法
骨架提取
raster map
indoor road network
refinement algorithm
skeleton extraction
