A methodology for the evaluation of street functions using video data: a case study on speed humps in Montreal

The direct observation of cars, pedestrians, cyclists, and other street users can be a viable method to evaluate the three main street functions, namely mobility, access, and place. However, a systematic procedure to evaluate the street functions is not evident in published work. Previously, a comprehensive framework for street functions and all users was proposed without any application. The aim of this research is therefore to develop a systematic methodology for collecting, pre-processing, and analyzing data on street users based on that comprehensive framework and to use it in a case study. In the proposed methodology, the trajectories and types of street users, their instantaneous speed, and direction of movement are automatically extracted from the collected videos using video analytics. These data are then analyzed in a new software tool, the Studio application, to derive street function evaluation indicators. The proposed method is applied to comprehensively assess the changes after speed hump installations in four residential streets in Montreal, Canada. The results demonstrate the value of direct street user observation and the proposed semi-automated method. The empirical results of the proposed method show that the speed of cars has decreased by 20-30% at all sites, while there have been significant changes in the flow and characteristics of vehicles, cyclists, and pedestrians in the study areas.

Mots clés

• street users
• street functions
• data collection
• video analysis
• indicator