NORMA eResearch @NCI Library

A Deep Learning Framework to Traffic Sign Recognition in All Weather Conditions

Piralkar, Aniket (2022) A Deep Learning Framework to Traffic Sign Recognition in All Weather Conditions. Masters thesis, Dublin, National College of Ireland.

[thumbnail of Master of Science]
PDF (Master of Science)
Download (3MB) | Preview
[thumbnail of Configuration manual]
PDF (Configuration manual)
Download (3MB) | Preview


Traffic Sign Recognition (TSR) for Advanced Driver Assistance Systems (ADAS) is crucial for preventing fatal crashes. However, recognizing the traffic signs in adverse weather and lighting conditions is a challenge. The aim of this research is to develop a deep learning framework to recognize the traffic signs in all weather conditions so that drivers can take appropriate action at the right time to avoid a fatal crash. An optimal deep learning model with a classification network will be developed to recognize relevant features of a particular traffic sign and classify it into appropriate categories. A German Traffic Sign Recognition dataset with 43 classes and around 50,000 images is used. This study illustrates a novel approach to classify traffic signs using the ResNet deep learning model and data augmentation. The performance of this model has achieved an accuracy of 98%. In terms of accuracy, this approach successfully classifies traffic signs in all weather conditions. It has also performed well compared to previous work in this field. The performance of the model is evaluated by using various metrics, including accuracy, mean average precision, and classification report. This is the critical step in the development of the deep learning model. However, future work can improve the accuracy further while maintaining current parameters.

Item Type: Thesis (Masters)
Uncontrolled Keywords: TSR; ADAS; ResNet; Data Augmentation; Residual Network; Deep Learning
Subjects: Q Science > QA Mathematics > Electronic computers. Computer science
T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science
T Technology > TE Highway engineering. Roads and pavements
T Technology > TL Motor vehicles. Aeronautics. Astronautics
Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Machine learning
Divisions: School of Computing > Master of Science in Data Analytics
Depositing User: Tamara Malone
Date Deposited: 28 Feb 2023 17:41
Last Modified: 01 Mar 2023 17:46

Actions (login required)

View Item View Item