Authors : Nilesh J. Uke and Ravindra C. Thool
Abstract: As web and digital cameras have become cheaper and wide-spread, the number of applications using vision techniques has increased significantly. Traffic detection (vehicles) and classification is one such application based on vision system. In this study, we propose a system for detection, classification and counting of vehicles on highway or expressway without explicit camera calibration. Frames taken from camera are processed to detect moving object on a logical line which enables counting of vehicles. Our experiments on several real traffic video sequences demonstrate good results for our foreground object detection, tracking, and classifying vehicles as light vehicles, heavy vehicles and motorcycles.
Nilesh J. Uke and Ravindra C. Thool, 2020. Images Analysis for Classification and Counting of Vehicles on Roads using OpenCV. International Journal of Signal System Control and Engineering Application, 13: 95-98.