Open AccessScene‐based pedestrian safety performance model in mixed traffic situation
Author(s) -
Zhang Yingying,
Yao Danya,
Qiu Tony Z.,
Peng Lihui
Publication year - 2014
Publication title -
iet intelligent transport systems
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.579
H-Index - 45
eISSN - 1751-9578
pISSN - 1751-956X
DOI - 10.1049/iet-its.2013.0012
Subject(s) - pedestrian , collision , computer science , transport engineering , traffic conflict , pedestrian detection , artificial intelligence , engineering , computer security , traffic congestion , floating car data
Compared to vehicle‐only collisions, traffic collisions with pedestrians are studied less because of insufficient data. However, with the development of image processing technology, a growing number of road user behavioural analyses have been conducted using video data. This study tries to extract road users’ movements from video data in order to analyse the conflict between pedestrian and vehicle and to evaluate pedestrian safety performance during conflicts. The time difference to collision (TDTC) parameter is used to fit the safety analysis on pedestrian‐involved conflicts. Scene‐based analysis, which evaluates the safety performance of traffic intersections and segments where several pedestrian–vehicle conflicts may happen together, is conducted using 91 groups of scene data. The parameters most related to pedestrian safety are located using a sensitivity test, a quantitative definition of pedestrian–vehicle conflict is then defined, and a scene‐based pedestrian safety performance evaluation model is built. The model can correctly detect nearly 94.4% of possibly dangerous traffic scenes. Other kinds of mixed traffic scenes can also be studied based on this research.