A New Similar Trajectory Retrieval Scheme Using k-Warping Distance Algorithm for Moving Objects

A k-warping distance algorithm is proposed which enhances the existing time warping distance algorithm by permitting up to k replications for an arbitrary motion of a query trajectory so that the authors measure the similarity between two trajectories accurately.

Abstract

In this paper, we propose a new similar trajectory retrieval scheme for efficient retrieval on both a single trajectory of a moving object and multiple trajectories of two or more moving objects. Our similar trajectory retrieval scheme can support multiple properties including direction, distance, and time and can provide the approximate matching that is superior to the exact matching. For this, we propose a k-warping distance algorithm which enhances the existing time warping distance algorithm by permitting up to k replications for an arbitrary motion of a query trajectory so that we measure the similarity between two trajectories accurately. In addition, we show from our experiment that our similar trajectory retrieval scheme using the k-warping distance algorithm outperforms Li's (no-warping) and Shan's schemes (infinite-warping) in terms of precision and recall measures. Finally, we implement a content-based soccer video retrieval system in order to show the usefulness of applying our similar trajectory retrieval scheme to a real application.