Hossain Bhuian, Dr. Mohammed BelalChowdhury, Faizul BariFaruqui, Tahmid AzimRazzaque, Rylah MarziaIqbal, Nayeem2024-01-162024-01-1620232023-01ID: 16321168ID: 17121072ID: 17121079ID: 181221047http://hdl.handle.net/10361/22162Cataloged from PDF version of final year design project.Includes bibliographical references (pages 74-79).This final year design project is submitted in partial fulfillment of the requirements for the degree of Bachelor of Science in Electrical and Electronic Engineering, 2023.Intelligent traffic system is not new to this modern world in order to mitigate traffic congestion. However, a proper management plan for a four way intersection lacks in our cities. An enhanced and optimized system is introduced in our project which in measure the density of all the vehicles plying on the road. The project is conducted on the Dhaka city where live traffic data has been collected from Shoinik Club Mor. A deep learning system is conducted where optimized YOLOv5 algorithm has been used to count the vehicles and the datasets for different vehicles are collected from COCO datasets and some datasets from BRTA (Bangladesh Road Transport Authority). After analyzing in different traffic situation for over 1260 frames at a rate of 30 FPS, we get around 88% accuracy in terms of detection. This operation sets the initial time for green signal as 16 seconds and then it changes according to the density occupied by vehicles of a particular road.91 pagesenBrac University project reports are protected by copyright. They may be viewed from this source for any purpose, but reproduction or distribution in any format is prohibited without written permission.Artificial intelligenceDeep learningYOLOv5DatasetsIoUNon max spressionImage processing -- Digital techniques.Cognitive learning theory (Deep learning)Project Report