Machine vison-based system for vehicle classification and counting using YOLO

• Main Authors: Refaie, Elbaraa, Mohd. Faudzi, Ahmad Athif
• Format: Book Section
• Published: Springer Science and Business Media Deutschland GmbH 2022
• Subjects: TK Electrical engineering. Electronics Nuclear engineering
• Online Access: http://eprints.utm.my/id/eprint/100670/; http://dx.doi.org/10.1007/978-981-16-8484-5_37

This paper proposes a machine vision-based system that is used for vehicle classification and counting. Vehicle classifier and counter are a very crucial in road design to determine the road load capacity. It is also important to monitor, manage, and analyze the traffic flow. In this work, the input video is obtained from a static video camera and from drone-captured video to count the cars, determine its direction and to classify the vehicle types on the road. The data will be fed to the vision-based system that will pass the frames captured by the camera through a YOLO neural network to classify the vehicles. Then, the cars contours are used to determine the car path that determines in which direction the vehicle is moving before it is being counted. The system will display the total number of vehicles moved in each direction and while identifying each vehicle’s type. This work is possible to be used for other application such as surveillance tracking and monitoring.