A semi-supervised learning framework based on spatio-temporal semantic events for maritime anomaly detection and behavior analysis

Abstract : Detection of abnormal movements of mobile objects has recently received a lot of attention due to the increasing availability of movement data and their potential for ensuring security in many different contexts. As timely detection of these events is often important, most current approaches use automated data-driven approaches. While these approaches have proved to be effective in specific contexts, they are not easily accepted by operators in charge of surveillance due, among other reasons, to the lack of user involvement during the detection process. To improve the detection and analysis of maritime anomalies this paper explores the potential of spatial ontologies for modeling maritime operator knowledge. The goal of this research is to facilitate the integration of human knowledge by modeling it in the form of semantic rules to improve confidence and trust in the anomaly detection system.