Spatio-temporal Similarity Measure for Network Constrained Trajectory Data

Abstract

Trajectory similarity measure is an important issue for analyzing the behavior of moving objects. In this paper, a similarity measure method for network constrained trajectories is proposed. It considers spatial and temporal features simultaneously in calculating spatio-temporal distance. The crossing points of network and semantic information of trajectory are used to extract the characteristic points for trajectory partition. Experiment results show that the storage space is decreased after trajectory partition and the similarity measure method is valid and efficient for trajectory clustering.