Analysis and detection system of accident using the surveillance cameras
Jisa Shaji, E.D. Dileesh · AIP conference proceedings · 2024
Nowadays, we are unaware of the safety of driving on roads.Accident cases have been increasing day by day.For a long period of time, accidents have been the main cause of road casualties.The city's level of traffic management is improved by the large-scale traffic flow system by its prediction and analysis.In early days, for security purposes, the traffic surveillance cameras have been installed.Hence, it requires manpower to survey each and every road's video streams.The dataset used here isthe video streams from the surveillance cameras that automatically analyse and detect the accident conditions.This project aims todetect the probability of an accident occurring or not occurring in the surveillance videos.As these surveillance videos are being recorded, they can be reverted directly to emergency services without any obtrusion.This could save the time wasted by the manual communication, as every second is important to saving lives during an accident.The paper proposes how human life on road can be simplified by using extraction techniques on the surveillance video.