基于PB-DBSCAN的GPS数据去噪  

GPS data denoising based on PB-DBSCAN

作　　者：汪鹏 刘泽玲[1] 王利琴 董永峰 WANG Peng;LIU Ze-ling;WANG Li-qin;DONG Yong-feng(School of Artificial Intelligence,Hebei University of Technology,Tianjin 300401,China;Hebei Key Laboratory of Big Data Computing,Hebei University of Technology,Tianjin 300401,China;Data Driven Industrial Intelligent Hebei Engineering Research Center,Hebei University of Technology,Tianjin 300401,China)

机构地区：[1]河北工业大学人工智能与数据科学学院,天津300401 [2]河北工业大学河北省大数据计算重点实验室,天津300401 [3]河北工业大学河北省数据驱动工业智能工程研究中心,天津300401

出　　处：《计算机工程与设计》2021年第3期678-683,共6页Computer Engineering and Design

基　　金：国家自然科学基金项目(61806072);天津市自然科学基金重点基金项目(19JCZDJC40000)。

摘　　要：针对公交车GPS数据量大、数据密度不均匀、噪声点多等问题,提出PB-DBSCAN(pixel_based-DBSCAN,PB-DBSCAN)算法。将聚类过程中判断数据点之间的关系改为判断像素格之间的关系,减小数据点邻域中的搜索范围,加快聚类速度。因公交线路的多样性,同一聚类参数无法适应所有线路,提出一种动态参数选择的方法。在石家庄公交车GPS实际数据集上进行实验,其结果表明,PB-DBSCAN可以有效识别并过滤GPS数据集中的噪声点,实现快速聚类。与采用固定参数的算法进行比较,参数的动态选择提高了聚类准确度。Aiming at the problems of large amount of GPS data,uneven data density and many noise points,PB-DBSCAN(pixel_based-DBSCAN,PB-DBSCAN)algorithm was proposed.The search range in the neighborhood of the data points was reduced by changing the relationship between the data points during the clustering process to the relationship between the pixel grids,and the clustering was speeded up.Due to the diversity of bus routes,the same clustering parameters could not adapt to all lines,and a dynamic parameter selection method was proposed.Experiments were performed on the Shijiazhuang bus GPS actual data set.The results show that PB-DBSCAN can effectively identify and filter noise points in the GPS data set,and can achieve fast clustering.Compared with the algorithm using fixed parameters,the dynamic selection of parameters improves the accuracy of clustering.

关 键 词：GPS轨迹数据 基于像素格的快速密度聚类 动态参数选择 像素格 去噪 

分 类 号：TP399[自动化与计算机技术—计算机应用技术]