基于大数据的船舶活动轨迹规律挖掘方法
Mining Method of Ship Activity Trajectory Pattern Based on Big Data
安健鹏 1李海霞 1雷亚丽 1王永权 1姚陈芳2
作者信息
- 1. 北方自动控制技术研究所,太原 030006
- 2. 战略支援部队中部预备役信息通信大队,太原 030000
- 折叠
摘要
针对当前船舶轨迹聚类技术存在特征属性研究单一的问题,提出一种基于融合距离的多维度船舶轨迹聚类算法技术,该技术通过加入多个属性特征并采用新的距离度量算法,从时序性和复杂度两方面提出了新的解决思路.在聚类结果基础上,针对缺少轨迹规律特征刻画方法的问题,提出基于局部区域均值的典型轨迹算法技术,通过对各属性进行均值计算,实现同类轨迹集合中轨迹特征的具体描绘.
Abstract
Aiming at the problem of single feature attribute research in current ship trajectory clustering technology,a multi-dimensional ship trajectory clustering algorithm based on merge distance is proposed.By adding multiple attribute features and adopting a new distance measurement algorithm,this technique provides a new solution idea from two aspects of time sequence and complexity.On the basis of clustering results,a typical trajectory algorithm based on local regional mean is proposed to solve the problem of the lack of trajectory feature characterization methods.By calculating the mean value of each attribute,the trajectory features in the same trajectory clustering are depicted in detail.
关键词
船舶轨迹聚类/相似性度量/典型轨迹提取/轨迹规律挖掘Key words
ship trajectory clustering/similarity measurement/typical trajectory extraction/trajec-tory pattern mining引用本文复制引用
出版年
2024
国家科技期刊平台
NSTL
万方数据