Clustering algorithm of vehicle motion trajectories in entrances and exits of freeway

DOI

10.2991/iccahe-16.2016.72How to use a DOI?

Keywords

Freeway entrances and exits; Trajectories analysis; Hausdorff distance; Clustering algorithm;

Abstract

In order to improve the research level of the intelligent traffic surveillance system, this paper presents a new hierarchical trajectory clustering algorithm in view of the features of trajectories in entrances and exits of freeway. The validity of trajectories is judged by means of the length of trajectories and vehicle speed firstly, and then the valid ones are encoded. The improved Hausdorff distance is proposed and applied to measure the similarity of temporal and spatial characteristics of trajectories. The fuzzy C-means hierarchical clustering algorithm of trajectories is further established in which trajectories are first clustered into different paths according to the position and geometric shape of trajectories, and then trajectories belonging to the same path are further clustered according to vehicle speed. Finally, the results of experiments in true scene have confirmed the validity and applicability of the algorithm.

Copyright

© 2016, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Zongyuan Sun
AU  - Dongxue Li
PY  - 2016/10
DA  - 2016/10
TI  - Clustering algorithm of vehicle motion trajectories in entrances and exits of freeway
BT  - Proceedings of the 2016 5th International Conference on Civil, Architectural and Hydraulic Engineering (ICCAHE 2016)
PB  - Atlantis Press
SP  - 421
EP  - 429
SN  - 2352-5401
UR  - https://doi.org/10.2991/iccahe-16.2016.72
DO  - 10.2991/iccahe-16.2016.72
ID  - Sun2016/10
ER  -