부산대학교 도서관

A traffic surveillance prevention system with vehicle number plate identification and speed detection using machine learning provides a complete traffic solution. By using only camera captures, a traffic observation system can measure major traffic boundaries from video layouts. This paper proposes to identify over-speeding vehicles' number plates and initiate an emergency message on detecting accidents. Utilizing the CCTV footage, vehicle detection takes place and the speed of the vehicle is derived using Open CV. The Number of the vehicle is recognized using contour under computer vision which is a field of machine learning that enables a computer system to derive meaningful information from digital images. The accident will be detected from the video input using the Convolutional Neural Network Algorithm and Computer Vision. On detecting an accident, an emergency message is initiated to the nearest control room with the corresponding location.