A Semantics-Based Trajectory Segmentation Simplification Method

Abstract With the development of mobile positioning technology, a large amount trajectory data has been generated. Therefore, to store, process and mine in better way, simplification is imperative. Current methods are either based on spatiotemporal features or semantic features, such as road network structure, but they do not consider stop. To overcome this limitation, study presents segmentation method stop features. The proposed first extracts feature trajectory, then divides into move segments finally simplifies obtained segments. verified by experiments personal taxi data. Compared with classic method, higher accuracy under different scales. especially suitable for more