Traffic Signal Classification using Deep Learning for Enhanced Autonomous Driving Systems

Prof. Zeba  Vishwakarma; Prof. Divya  Pandey; Prof. Mallika  Dwivedi; Arjita  Sarkar; Gourav  Mourya

doi:10.15662/IJAREEIE.2022.1204045

Prof. Zeba Vishwakarma Department of Computer Science and Engineering, Baderia Global Institute of Engineering and Management, Jabalpur, MP, India
Prof. Divya Pandey Department of Computer Science and Engineering, Baderia Global Institute of Engineering and Management, Jabalpur, MP, India
Prof. Mallika Dwivedi Department of Computer Science and Engineering, Baderia Global Institute of Engineering and Management, Jabalpur, MP, India
Arjita Sarkar Department of Computer Science and Engineering, Baderia Global Institute of Engineering and Management, Jabalpur, MP, India
Gourav Mourya Department of Computer Science and Engineering, Baderia Global Institute of Engineering and Management, Jabalpur, MP, India

DOI: https://doi.org/10.15662/IJAREEIE.2022.1204045

Keywords: Traffic Signal Classification, Deep Learning, Autonomous Driving, Convolutional Neural Network, Driver Alert Systems, Hyper-parameter Tuning

Abstract

This paper presents a deep learning-based model for the classification of traffic signals, a crucial component for the development of autonomous driving systems and driver alert mechanisms. Utilizing a convolutional neural network (CNN), the proposed model is designed to recognize 43 distinct traffic signs with high accuracy. The model is trained on a publicly available dataset, achieving an impressive accuracy of over 95% in just 50 epochs. The preprocessing steps, model architecture, and training process are discussed in detail. The model's performance indicates its potential for real-world applications in enhancing the safety and efficiency of self-driving cars. Future work includes hyper parameter tuning and the integration of the model into real-time systems for further optimization.