Surface traffic monitoring using OpenCV for various weather conditions through enhanced spatial correlation method
Dipak C. Ghosh, Soumokanti Bera, Pushan Kumar Datta · 2022 International Mobile and Embedded Technology Conference (MECON) · 2022
It is a big challenge in the present urban traffic management system to develop a vehicle detection and traffic surveillance monitoring system through video capture using real time data capture model whereby machine learning may be used to recognise and categorise things. The idea is to analyse the motion and approach of cars acquired from CCTV cameras footage installed on the roads in different weather conditions and create an interactive software which should be able to gather and update important statistics like vehicle type classification and counts, lane usage details, for better regulation and control with ease. The core of this working model is to enhance vehicle detection and counting module which is implemented by analysing consecutive frames of video using frame differencing technique. The government has tried the best to ensure swift movement by implementing the broad usage of intelligent traffic controller system design. According to the WHO, around 2 million road traffic accidents result in deaths, with some nations devoting 2% of their GDP to road traffic accidents. Opportunities in the intelligent transport system market is growing at a global compound annual growth rate (CAGR) from 2019 to 2025 as an effective tool for road safety.