| With the process of urbanization in China,the number of motor vehicles and the density of population are increasing rapidly.The urban traffic problem is becoming more and more serious.As a vulnerable group of roads,pedestrians are often more vulnerable in the mixed road trafic environment.According to statistics,pedestrian crashes account for more than 20%of the traffic accidents in urban roads every year,and have a high proportion of casualties.Therefore,it is necessary to carry out the research on the distribution and influencing factors of urban pedestrian crashes,and put forward the preventive countermeasures and suggestions.Based on the actual pedestrian crashes on urban roads,this paper analyzes urban pedestrian crashes based on geographic information systems.Through the actual pedestrian traffic accident location coding,the map display of pedestrian crashes can be realized,and then the kernel density is calculated;the distribution characteristics of urban pedestrian crashes are obtained by combining the kernel density distribution of pedestrian crashes with the kernel density distribution of population and road network,and several main black hot spots are identified and analyzed.At the same time,the pedestrian crashes database is expanded based on street view maps and other geographic information system resources,which further enriches the accident information.On this basis,the main factors affecting the severity of pedestrian injury in pedestrian crashes are obtained by combining the binary logistic regression model and the tree-based model.And the empirical analysis is carried out,and the countermeasures and measures to reduce pedestrian injuries in pedestrian crashes are put forward.The main research results of this article are as follows:(1)The integration of traditional traffic accident data and rich geographic information system resources further expands the accident data,which provides conditions for discussing the spatial distribution and influencing factors of urban road pedestrian crashes from the four aspects of people,vehicles,roads and environment in macro and micro.(2)Through geographic location coding and network-based kernel density calculation,the distribution characteristics of urban pedestrian crashes are obtained:pedestrian crashes present several cluster distributions in urban space,and are significantly related to population density,road network structure,and social services and activities.(3)According to the visualization results of geographic space and the predictive analysis of the model,pedestrian age,light conditions,road physical isolation,road type,time period,pedestrian walking speed,etc.have become important factors affecting pedestrian injury severity in pedestrian crashes.This study not only provides a reference for us to effectively avoid and reduce traffic injuries in real life,but also provides a basis for urban traffic management departments to develop targeted measures to improve the safety environment of pedestrians,and provides a basis reference for relevant experiments and study of active and passive safety for pedestrian protection. |