United States Department of Transportation
ROSA P serves as an archival repository of USDOT-published products including scientific findings, journal articles, guidelines, recommendations, or other information authored or co-authored by USDOT or funded partners.
As a repository, ROSA P retains documents in their original published format to ensure public access to scientific information.
i
A Comparative Study of Pedestrian Crossing Behavior and Safety in Baltimore, MD and Washington, DC Using Video Surveillance
-
2023-08-01
[PDF-2.51 MB]
Details:
-
Creators:
-
Corporate Creators:
-
Contributors:
-
Corporate Contributors:
-
Subject/TRT Terms:
-
Resource Type:
-
Geographical Coverage:
-
Edition:Final Report
-
Corporate Publisher:
-
Abstract:Pedestrian safety is of paramount importance in urban transportation, and it is a key goal of initiatives such as Vision Zero. Washington, D.C., and Baltimore, Maryland, have experienced a significant number of traffic accidents in recent years, with pedestrians being particularly vulnerable. Video surveillance has long been established as a valuable method for analyzing pedestrian behavior. However, traditional manual analysis of video footage is costly, time-consuming, and prone to human errors. With the advancement of computer vision and machine learning technologies, such as the YOLO (You Only Look Once) algorithm, it has become possible to automate and streamline the analysis process. In this study, we applied theYOLOv8 algorithm to analyze video surveillance footage, allowing for efficient extraction of pedestrian data and the development of analytics for pedestrian behavior at signalized intersections. By leveraging computer vision and machine learning tools, we were able to process large volumes of video footage and obtain detailed insights into pedestrian behavior at intersections. The application of computer vision and machine learning techniques, specifically the YOLOv8 algorithm, to analyze video surveillance footage has proven to be a valuable approach for studying pedestrian behavior at intersections in Washington, D.C., and Baltimore, Maryland. This preliminary study presents a method for tracking pedestrians and compares pedestrian volume and speed across five intersections. By automating the analysis process, this study has provided comprehensive insights for using video footage to track pedestrian movements, contributing to the broader goal of improving pedestrian safety in urban environments. Future research endeavors should develop additional algorithms to improve pedestrian tracking and provide additional insights to vehicle and pedestrian interaction to enhance pedestrian safety.
-
Format:
-
Funding:
-
Collection(s):
-
Main Document Checksum:
-
Download URL:
-
File Type:
