多维密度聚类的精细化道路交通运行状况检测  被引量:7

Refined Traffic State Detection of Road Based on Multidimensional Density Clustering

在线阅读下载全文

作  者:孙梦婷 魏海平[1] 李星滢 于靖宇 徐立[1] SUN Mengting;WEI Haiping;LI Xingying;YU Jingyu;XU Li(Information Engineering University,Zhengzhou 450001,China)

机构地区:[1]信息工程大学

出  处:《测绘科学技术学报》2019年第4期412-417,共6页Journal of Geomatics Science and Technology

摘  要:现有的路况检测方法以整条路段为单位进行检测,存在精度不高的问题,且DBSCAN算法用于出租车GPS数据聚类仍存在脱离线性参照系统、假噪声和簇内速度差异大等问题。在线性参照系统中定位GPS点,以两点间的测量值距离作为空间距离,同时增加速度距离约束,提出一种基于DBSCAN算法的多维密度聚类算法,使其适用于精细化路况检测;在此基础上构建路况事件表,并利用动态分段技术对路况事件进行管理和可视化,满足实际应用中对路况检测精度的要求。以上海市出租车GPS数据和路网数据为例进行实验分析,结果表明,提出的方法能够实现较为精细的路况检测。The accuracy of detection is not high enough because the existing road condition detection methods take the whole section as the unit to carry on,and there are still some problems with DBSCAN algorithm for taxi GPS data clustering such as breaking away from linear reference system,false noise and big difference in cluster speed.Measuring GPS points in the linear reference system and taking the measured value distance between two points as the spatial distance,and at the same time increasing the speed distance constraint,then a multi-dimensional density clustering algorithm based on DBSCAN algorithm was proposed to make it suitable for the refined road condition detection.On this basis,the road condition events table was constructed to manage and visualize the road condition events by using the dynamic segmentation technology.And the requirement of road condition detection accuracy in practical application has been solved.Taking the taxi GPS data and road network data of Shanghai as an example,the results show that the proposed method can realize the refined road condition detection.

关 键 词:出租车GPS数据 多维密度聚类 路况检测 动态分段技术 路况事件表 

分 类 号:P208[天文地球—地图制图学与地理信息工程]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象