基于关键区域的井下人员轨迹挖掘方法
|
摘要
提出了基于关键区域的井下人员轨迹挖掘框架,该框架由关键位置发现算法和移动对象轨迹挖掘算法组成。首先利用关键位置发现算法将矿井下的定位数据转化成有特定语义的关键位置序列;然后利用移动对象轨迹挖掘算法将关键位置序列聚类关键区域,从而发现井下移动对象的日常轨迹,之后利用轨迹结构相似度筛选出异常轨迹。利用矿工定位数据集进行试验表明:基于关键区域的井下人员轨迹挖掘框架解决了多密度区域的识别问题,能够准确识别出矿工日常轨迹和异常轨迹。
A trajectory mining framework for underground personnel based on key areas is proposed. The framework consists of key location discovery algorithm and moving object trajectory mining algorithm. Firstly, the key location discovery algorithm is used to transform the positioning data under the mine into a key position sequence with specific semantics. Then the moving object trajectory mining algorithm is used to cluster the key position sequences into key regions, so as to find the daily trajectory of the moving objects in the underground, and then use the trajectory. Structural similarity screens out anomalous trajectories.Experiments using miners' positioning datasets show that the trajectory mining framework based on key areas solves the problem of multi-density area identification, which can accurately identify the daily trajectory and abnormal trajectory of miners.
A trajectory mining framework for underground personnel based on key areas is proposed. The framework consists of key location discovery algorithm and moving object trajectory mining algorithm. Firstly, the key location discovery algorithm is used to transform the positioning data under the mine into a key position sequence with specific semantics. Then the moving object trajectory mining algorithm is used to cluster the key position sequences into key regions, so as to find the daily trajectory of the moving objects in the underground, and then use the trajectory. Structural similarity screens out anomalous trajectories.Experiments using miners' positioning datasets show that the trajectory mining framework based on key areas solves the problem of multi-density area identification, which can accurately identify the daily trajectory and abnormal trajectory of miners.
引文
[1]佚名.国务院关于进一步加强企业安全生产工作的通知[J].中国应急管理,2010(7):9-13.
[2] Luis O A, Vania B, Bart K, et al. A model for enrich-ing trajectories with semantic geographical information[C]∥Proceedings of the 15th Annual ACM Interna-tional Symposium on Advances in Geographic Informa-tion Systems.New York:ACM, 2007.
[3]袁冠.移动对象轨迹数据挖掘方法研究[D].徐州:中国矿业大学,2012.
[4]吕绍仟,孟凡荣,袁冠.基于轨迹结构的移动对象热点区域发现[J].计算机应用,2017,37(1):54-59.
[5]徐尽,田胜利.基于拐点判断法的GPS定位数据精简[J].计算机工程,2010,36(7):268-269.
[6]郑宇,谢幸.基于用户轨迹挖掘的智能位置服务[J].中国计算机学会通讯,2010,6(6):23-30.
[7]刘畅,李治军,姜守旭.基于DBSCAN算法的城市交通拥堵区域发现[J].智能计算机与应用,2015,5(3):69-71.
[8]周培培,丁庆海,罗海波,等.基于DBSCAN聚类算法的异常轨迹检测[J].红外与激光工程,2017(5):230.
[9]朱燕,李宏伟,樊超,等.基于聚类的出租车异常轨迹检测[J].计算机工程,2017,43(2):16-20.
[10]张晓滨,杨东山.基于时间约束的Hausdorff距离的时空轨迹相似度量[J].计算机应用研究,2017,34(7):2077-2079.
[2] Luis O A, Vania B, Bart K, et al. A model for enrich-ing trajectories with semantic geographical information[C]∥Proceedings of the 15th Annual ACM Interna-tional Symposium on Advances in Geographic Informa-tion Systems.New York:ACM, 2007.
[3]袁冠.移动对象轨迹数据挖掘方法研究[D].徐州:中国矿业大学,2012.
[4]吕绍仟,孟凡荣,袁冠.基于轨迹结构的移动对象热点区域发现[J].计算机应用,2017,37(1):54-59.
[5]徐尽,田胜利.基于拐点判断法的GPS定位数据精简[J].计算机工程,2010,36(7):268-269.
[6]郑宇,谢幸.基于用户轨迹挖掘的智能位置服务[J].中国计算机学会通讯,2010,6(6):23-30.
[7]刘畅,李治军,姜守旭.基于DBSCAN算法的城市交通拥堵区域发现[J].智能计算机与应用,2015,5(3):69-71.
[8]周培培,丁庆海,罗海波,等.基于DBSCAN聚类算法的异常轨迹检测[J].红外与激光工程,2017(5):230.
[9]朱燕,李宏伟,樊超,等.基于聚类的出租车异常轨迹检测[J].计算机工程,2017,43(2):16-20.
[10]张晓滨,杨东山.基于时间约束的Hausdorff距离的时空轨迹相似度量[J].计算机应用研究,2017,34(7):2077-2079.