Traffic Signs Classification using Convolutional Neural Network (CNN)

Vidisha Verma, Aashna Khan, Shailendra Tiwari

2021

Abstract

A complete ATSR system is proposed in this paper using a Convolutional Neural Network (CNN) based classifier. After several assessments, a seven-layer architecture has been implemented that was designed to be fast enough to serve a particular application. The training process was carried out with the Adam’s Optimization algorithm as an alternate to the Stochastic Gradient Descent (SGD) method. The widely known “German Traffic Sign Dataset – GTSRB” was partitioned into training, testing and validation datasets. The development process is delineated in the paper, and it displays the pipeline utilized in the image processing. The technique put forward was accurate in determining 96.35% of the validation data with a significantly smaller and faster system compared to similar models

Paper Citation

in Harvard Style

Verma V., Khan A. and Tiwari S. (2021). Traffic Signs Classification using Convolutional Neural Network (CNN). In Proceedings of the 3rd International Conference on Advanced Computing and Software Engineering - Volume 1: ICACSE, ISBN 978-989-758-544-9, pages 212-217. DOI: 10.5220/0010567100003161

in Bibtex Style

@conference{icacse21,
author={Vidisha Verma and Aashna Khan and Shailendra Tiwari},
title={Traffic Signs Classification using Convolutional Neural Network (CNN)},
booktitle={Proceedings of the 3rd International Conference on Advanced Computing and Software Engineering - Volume 1: ICACSE,},
year={2021},
pages={212-217},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0010567100003161},
isbn={978-989-758-544-9},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 3rd International Conference on Advanced Computing and Software Engineering - Volume 1: ICACSE,
TI - Traffic Signs Classification using Convolutional Neural Network (CNN)
SN - 978-989-758-544-9
AU - Verma V.
AU - Khan A.
AU - Tiwari S.
PY - 2021
SP - 212
EP - 217
DO - 10.5220/0010567100003161