Experimental study of pedestrian behaviors in a corridor based on digital image processing

Evacuation experiments of pedestrians passing through a corridor, which acts as a bottleneck, are conducted in this study. A video processing method is used to extract the trajectories of pedestrians in the measurement section. The velocity, density, flow rate and time headway between successive pedestrians entering the bottleneck are calculated, and the relations between them are obtained from the trajectories. We report that the mean velocities for each experiment vary greatly, depending on the lanes contained in the measurement section. The mean velocity of each pedestrian in the corridor displays a trend that decreases sharply at the beginning and then fluctuates slightly around the mean value. A fundamental diagram relating density and speed is obtained. We also investigate the relation between the mean velocities of pedestrians and the density, as a function of time, and report that the mean velocity decreased while the density increased. These results are consistent with those of other researchers. The flow rate increases linearly with the bottleneck width, which is consistent with other results. The time headways are investigated in this study. It is observed that the behavior of pedestrians in the corridor, such as when they formed lanes, has a significant effect on the time headways and their distribution. The distribution of time headways in different lanes is also studied, and some interesting results are obtained. The data and results provided in this paper may be useful for improving evacuation models and designing public facilities. (C) 2011 Elsevier Ltd. All rights reserved.