Automatic Traffic Monitoring and Management for Pedestrian and Cyclist Safety Using Deep Learning and Artificial Intelligence

University Transportation Centers Program2020PDFTech ReportPourhomayoun, MohammadMineta Consortium for Transportation MobilityUnited States. Department of Transportation. Office of the Assistant Secretary for Research and TechnologySan Jose State University. College of BusinessUnited States. Department of Transportation. University Transportation Centers (UTC) ProgramMineta Transportation InstituteUnited StatesArtificial intelligenceMachine learningPedestrian countsPedestrian safetyTraffic safety20-2965A0660Final ReportIn this project, we have designed and developed an effective end-to-end system based on advanced Artificial Intelligence (AI), machine learning, and computer vision to automatically monitor, detect, track, and count pedestrians and bicyclists. The main objective of this project is to improve the safety of pedestrians and bicyclists, by applying self-sensed and AI-powered systems to monitor and control the flow of pedestrians/bicyclists. The developed system includes algorithms for detecting the pedestrians and bicyclists, as well as algorithms for tracking and counting the pedestrians. We evaluated the developed system on real videos captured by actual traffic cameras in the city of Los Angeles. Despite the low quality of some of the videos, the results demonstrated high accuracy and effectiveness of the developed system in automatically detecting and counting pedestrians and bicyclists.127