An interactive method for identifying the stay points of the trajectory of moving objects

摘要

The stay points of trajectory include important semantic information. Identifying the stay points is an indispensable step for mining and analyzing the trajectory data. Current methods for identifying stay points generally require pre-setting thresholds, which have a major impact on the identification results. Meanwhile, the setting of these thresholds is heavily reliant on empirical experience. There are many types of moving objects, such as vehicles, ships, airplanes, people and animals. These moving objects have a large variation in their trajectory data due to their different geometric/physical properties and navigation environments. Therefore, setting the appropriate thresholds for the identification process is by no means easy for non-professionals. To cope with this challenge, an interactive visualization method is suggested in this paper in order to help users to set the thresholds. In the proposed method, the Space-Time Cube (STC) is used to visualize the trajectory as the first step. In the next step, users interactively choose the typical stay points of the trajectory. The threshold is atomically determined using the geometric characteristics of the 3-dimensional bounding box of the space where the moving object stays on. Finally, the sliding window method is used to extract the stay points. Experimental results demonstrate that the proposed method is an efficient method for determining the thresholds in an intuitive and fast manner. As a result, it can significantly improve the efficiency of identifying the stay points of the trajectory for non-professionals. (C) 2019 Published by Elsevier Inc.