Traffic Sign Recognition Using Deep Learning: a Better Way to Safe Driving

EasyChair Preprint 12388

Abstract

The number of accidents caused by failure to observe traffic signs and follow traffic laws has been steadily growing. Using synthesised training data generated from road traffic sign photos, we may overcome the limitations of traffic sign detection databases, which differ between countries and regions. This technology is used to create a library of synthesised images for detecting traffic signs under various lighting situations. We can create a data-driven traffic sign identification and detection system with high detection accuracy and high-performance capabilities in training and recognition procedures using this data set and a perfect Convolutional Neural Network (CNN). This reduces the number of accidents and allows the driver to concentrate on driving rather than studying every traffic sign. The goal of this work is to present an effective approach for detecting and recognising traffic signs in India. We developed approaches such as neural networks and feature extraction to overcome the limits of existing methods, increase the effectiveness of recognising traffic signs, and minimise road accidents.

Keyphrases: Convolutional Neural Network, Road accidents, Traffic sign recognition, feature extraction

@booklet{EasyChair:12388,
  author    = {Rishub Kumar and Shekharesh Barik},
  title     = {Traffic Sign Recognition Using Deep Learning: a Better Way to Safe Driving},
  howpublished = {EasyChair Preprint 12388},
  year      = {EasyChair, 2024}}