Citation

Al Farid, Fahmid and Karim, Hezerul Abdul and Bhuiyan, Md Roman and Badie, Farshad and Balaganesh, Duraisamy and Tusher, Md. Mahbubur Rahman (2025) Traffic Sign Classification Across Borders: Evaluating DenseNet201 on Belgium and Bangladesh Signs. In: 2025 Multimedia University Engineering Conference (MECON), 21-23 July 2025, Cyberjaya, Malaysia.

Text 484.pdf - Published Version Restricted to Repository staff only Download (1MB)

Abstract

Traffic Sign Recognition (TSR) is essential for autonomous vehicles and intelligent transportation systems. This study compares deep learning-based TSR models trained on two datasets: Belgium Traffic Sign Classification (BTSC) with 62 classes, and Bangladesh Traffic Sign Recognition (BDTSR) with 48 classes. Using a DenseNet201 backbone pre-trained on ImageNet, we applied transfer learning with a custom classification head, data augmentation, and fine-tuning. The model achieved 95.34% validation accuracy on BTSC and 99.17% on BDTSR. The results demonstrate the effectiveness of transfer learning across regions, even with differing signage styles and environ- mental conditions. This highlights the importance of localized training data for building accurate TSR systems, helping improve road safety and autonomous navigation in diverse environments.

Item Type:	Conference or Workshop Item (Paper)
Uncontrolled Keywords:	Bangladeshi traffic signs, Belgium traffic signs, deep learning, DenseNet201, traffic sign recognition
Subjects:	T Technology > TL Motor vehicles. Aeronautics. Astronautics
Divisions:	Faculty of Artificial Intelligence & Engineering (FAIE)
Depositing User:	Ms Suzilawati Abu Samah
Date Deposited:	17 Mar 2026 04:06
Last Modified:	19 Mar 2026 01:39
URII:	http://shdl.mmu.edu.my/id/eprint/15476

Downloads

Edit (login required)