Traffic Sign Recognition & Detection using Transfer learning

Sharma, Akhil (2019) Traffic Sign Recognition & Detection using Transfer learning. Masters thesis, Dublin, National College of Ireland.

Preview

PDF (Master of Science) Download (2MB) | Preview

Abstract

Traffic sign recognition feature is widely employed in industry today by researchers working in artificial intelligence and machine learning fields targeted to create an autonomous driving assistance system. This will help the driver’s human eye and manoeuvring capabilities which can’t be presumed to be consistent for high accuracy for all weather condition challenges. As the limitation of the human eye perception on a road sign is upsetting the road safety. This research paper illustrates a novel approach for Traffic image classification using Deep learning model to distinguish the different classes of traffic signs only after applying Data augmentation offered by Keras deep learning library for shifting traffic signs images (5%,10%,15%) and applying Gaussian blur technique for blurring (3:3,5:5 & 7:7) images. This will help us re-create a dataset similar to images captured during bad climatic, foggy or windy conditions seen in Ireland or badly captured images due to fast moving cars with limitation in camera angles from roads. We will be introducing these pictures on a pre-trained transfer learning model VGG19 as it has better accuracy and operational speed to create a best fit model. Results of this research work can be used by automobile companies to create enhanced image capturing cameras and classification models to make better autonomous systems and safer roads.

Item Type:	Thesis (Masters)
Subjects:	Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science Q Science > QA Mathematics > Computer software T Technology > T Technology (General) > Information Technology > Computer software
Divisions:	School of Computing > Master of Science in Data Analytics
Depositing User:	Caoimhe Ní Mhaicín
Date Deposited:	11 Oct 2019 15:36
Last Modified:	11 Oct 2019 15:36
URI:	https://norma.ncirl.ie/id/eprint/3854

Actions (login required)

View Item