Multi-scale Classification of Moving Object Trajectories
Cédric du Mouza, Philippe Rigaux · HAL - CNAM · 2004
In this paper we propose a classification model for moving objectstrajectories. We assume that the classification is based on amulti-scale map, and we simply define a trajectory pattern asthe sequence of zones an object crosses during its travel. Thesepatterns constitute the basis of classification operators. We alsodefine a pattern-based query language which allows an online andcontinuous classification of moving objects. Finally a prototype whichshows the validity of the approach is briefly described.