Fast and parameter-light rare behavior detection in maritime trajectories
作者：
Highlights：
• A scheme cross trajectories, vessel attributes and the movement context for detecting rare behaviors through preprocessing, kNN-based clustering, and verification.
• Only extremely few anomalies are useful which are rare behaviors.
• The interactive detection process requires an instant response that is a big challenge for.
• The more similar trajectories gather in an Area of Interest, the less probability of anomalies they are.
• The small reachability distance values have limited effect on lrd.
摘要
•A scheme cross trajectories, vessel attributes and the movement context for detecting rare behaviors through preprocessing, kNN-based clustering, and verification.•Only extremely few anomalies are useful which are rare behaviors.•The interactive detection process requires an instant response that is a big challenge for.•The more similar trajectories gather in an Area of Interest, the less probability of anomalies they are.•The small reachability distance values have limited effect on lrd.
论文关键词：Visual analytics,Rare behaviour detection,Trajectory
论文评审过程：Received 3 January 2020, Revised 27 March 2020, Accepted 12 April 2020, Available online 24 May 2020, Version of Record 24 May 2020.
论文官网地址：https://doi.org/10.1016/j.ipm.2020.102268
