摘要
为了解决大多数的轨迹简化算法缺乏对轨迹多个特征的综合衡量的问题,提出了一种融合位置、方向、速度与时序等多个轨迹特征,基于阈值引导采样的船舶轨迹简化算法。该算法以角度阈值法思想为基础,计算轨迹点的前后多个特征差,与设置的阈值进行对比,进而实现对轨迹的简化。利用船舶AIS轨迹数据对该算法进行轨迹简化和轨迹聚类实验。实验结果表明,该轨迹简化算法在简化率、简化误差率上均优于角度阈值算法,能保留原始轨迹的形状,并且简化后的轨迹数据能运用于轨迹聚类,聚类效果良好。
In order to solve the lack of comprehensive measurement of multiple features of trajectory in most simplified algorithms,we propose a trajectory simplification algorithm which is based on the angle threshold algorithm and takes into account multiple trajectory features such as position,direction,speed and time sequence.The proposed algorithm calculates the differences between features of the successive trajectory points and compares them with the set thresholds to simplify the trajectories.Using ship s AIS trajectory data,trajectory simplification and trajectory clustering are performed on the proposed algorithm.The results of the experiments show that the proposed algorithm is superior to the angle threshold algorithm in terms of simplification rate and simplification error rate.The results also show that it can retain the shape of the original trajectory and the simplified trajectory can effectively be used for trajectory clustering.
作者
张银昊
潘家财
赵梦鸽
ZHANG Yinhao;PAN Jiacai;ZHAO Mengge(Navigation College,Jimei University,Xiamen 361021,China)
出处
《集美大学学报（自然科学版）》
CAS
2021年第5期425-432,共8页
Journal of Jimei University：Natural Science
关键词
船舶轨迹
简化算法
阈值引导采样法
船舶自动识别系统
移动对象
ship trajectory
simplification algorithm
threshold guide sampling algorithm
ship automatic identification system
moving object
