摘要
为解决传统在线压缩算法在渔船轨迹点密集处易丢失轨迹,难以保留方向变化较大的特征点的问题,提出一种基于改进滑动窗口的渔船轨迹在线压缩算法。该算法将船舶领域的思想与经典滑动窗口算法相结合,在滑动窗口中加入一个可变的椭圆形船舶领域作为判断压缩轨迹点的依据。实验结果表明,该算法能够在不影响算法运行时间的同时更好地保留渔船轨迹的局部特征,并且在不同阈值下的压缩率比经典滑动窗口算法的平均提高了约5%,压缩效果更优。
In order to solve the problem that the traditional online compression algorithm is easy to lose trajectories at the dense fishing ship trajectory points and it is difficult to retain the feature points with large direction changes,an online compression algorithm of fishing ship trajectories based on improved sliding window is proposed.The algorithm combines the idea of ship domain with the classical sliding window algorithm,and uses a variable elliptical ship domain in the sliding window as the basis for judging and compressing the trajectory points.The experimental results show that,the algorithm can better retain the local features of fishing ship trajectories without affecting the algorithm running time,and also improves the compression rate by about 5%on average than the classical sliding window algorithm under different thresholds,showing a better compression effect.
作者
顾杰
宋鑫
施笑畏
苌道方
GU Jie;SONG Xin;SHI Xiaowei;CHANG Daofang(Logistics Engineering College,Shanghai Maritime University,Shanghai 201306,China;Qingdao Institute,Shanghai Maritime University,Qingdao 266237,Shandong,China;Institute of Logistics Science&Engineering,Shanghai Maritime University,Shanghai 201306,China)
出处
《上海海事大学学报》
北大核心
2023年第4期17-24,共8页
Journal of Shanghai Maritime University
基金
工业和信息化部高技术船舶项目(MC-201917-C09)。
作者简介
顾杰(1998—),男,江苏南通人,硕士研究生,研究方向为AIS数据分析与处理、数据挖掘等,(E-mail)202030210302@stu.shmtu.edu.cn;施笑畏(1973—),女,浙江嘉兴人,副教授,博士,研究方向为人机工程应用、数据挖掘、个性化内容推荐等,(E-mail)xwshi@shmtu.edu.cn;苌道方(1978—),男,河南新乡人,教授,博士,研究方向为供应链与物流规划与设计、复杂系统建模与仿真、智能算法与工业APP开发,(E-mail)dfchang@shmtu.edu.cn。
