Wrong-way-vehicle-detection: Wrong-way driving is a significant contributor to road accidents and traffic congestion worldwide. By effectively identifying wrong-way vehicles, we can greatly reduce the number of accidents and alleviate traffic congestion. In this project, we present an automated system for detecting wrong-way vehicles using surveillance camera footage captured on the road. Our system operates in three stages: vehicle detection using the You Only Look Once (YOLOv5) algorithm, vehicle tracking within a specified region of interest using the centroid tracking algorithm, and identification of wrong-way driving vehicles. YOLO demonstrates high accuracy in object detection, while the centroid tracking algorithm efficiently tracks moving objects. Experimental results using various traffic videos demonstrate the effectiveness of our proposed system in detecting and identifying wrong-way vehicles under different lighting and weather conditions.  

Code Link: https://github.com/zillur-av/wrong-way-vehicle-detection

Paper Link: https://arxiv.org/abs/2210.10226