Smart Traffic Counting System

Abstract

Nowadays, the number of vehicles on the road is increasing rapidly, creating problems for effective traffic monitoring and control. Traditional traffic counting systems are not capable of keeping up with the growing volume and complexity of urban traffic, resulting in inaccuracies and delays while collecting the data. This thesis will focus on the application of machine vision and intelligent automation to create a smart traffic counting system.

This system takes an image and or any segment of the video, and it will analyze it using a trained model version of YOLO. It will identify and compare the vehicles using datasets from the model version, and processes the data in real time, detecting and counting the vehicles under the conditions specified by the user.

By using this system, operators and planners will be able to enhance the accuracy, efficiency, and reliability of traffic counting beyond traditional counting methods. This research will emphasize the potential of machine vision technology to transform traffic monitoring, allowing for better urban planning and well-structured advanced traffic systems.