Title :
T-junction: Experiments, trajectory collection, and analysis
Author :
Boltes, Maik ; Zhang, Jun ; Seyfried, Armin ; Steffen, Bernhard
Author_Institution :
Forschungszentrum Julich GmbH, Julich, Germany
Abstract :
For the proper understanding and modelling of pedestrian dynamics, reliable empirical data is necessary for analysis and verification. This paper discusses laboratory experiments at T-junctions: the setup, the extraction of the trajectories of the pedestrians and the analysis of the resulting data. Such experiments give us the opportunity to selectively analyse parameters independent of undesired influences and adjust them to high densities seldom seen in field studies. For the T-junction the density inside the junction has been varied and analysed. Two strategies for the time-efficient automatic extraction of accurate pedestrian trajectories from stereo recordings are presented. One strategy uses marker for detection and the other one based on a perspective depth field. From these trajectories the fundamental diagram of T-junction flow is analysed, the density, velocity and specific flow profiles are obtained using the Voronoi method.
Keywords :
computational geometry; data analysis; feature extraction; image motion analysis; stereo image processing; T-junction flow; Voronoi method; laboratory experiments; pedestrian dynamic modeling; resulting data analysis; stereo recordings; time-efficient automatic pedestrian trajectory extraction; trajectory extraction;
Conference_Titel :
Computer Vision Workshops (ICCV Workshops), 2011 IEEE International Conference on
Conference_Location :
Barcelona
Print_ISBN :
978-1-4673-0062-9
DOI :
10.1109/ICCVW.2011.6130238