Traffic Vehicle Detection of Dhaka City using Deep Learning Algorithm

Abstract

The fast urbanization of Dhaka City has resulted in substantial traffic congestion, requiring effective management techniques. This study investigates the use of deep learning methods, namely convolutional neural networks (CNN), to detect automobiles in traffic in Dhaka. The CNN is trained using photos collected from several places throughout the city. The model is specifically designed to provide optimal performance in detecting objects in real-time, effectively overcoming the difficulties presented by the city's high population density and varied traffic conditions. Preprocessing techniques like picture augmentation and normalization improve the model's ability to handle different scenarios, and its performance is assessed using precision, recall, and F1-score measures. The results demonstrate that the deep learning model outperforms conventional methods in terms of both accuracy and speed, implying significant enhancements for traffic monitoring and management. This study highlights the capacity of deep learning in addressing urban traffic issues, hence facilitating the development of sophisticated intelligent transportation systems in Dhaka.