Deep Learning Networks for Detection, Classification and Analysis of Car Damage

Chaudhari, Shubham Sarjerao (2022) Deep Learning Networks for Detection, Classification and Analysis of Car Damage. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (1MB) | Preview
Preview	PDF (Configuration manual) Download (3MB) | Preview

Abstract

Cars are incredibly essential in the contemporary generation, and the ability to detect car damage autonomously is a feature that the automobile insurance industry is quite interested in. Assigning an object to each of these groups visually is known as classifying it. Machine learning as well as computer vision are used in visual image recognition technology. Research started by training a fundamental CNN model. Despite this, it fails to perform well due to a small and unbalanced data sample. The effect of model pretraining proceeded by fine model tuning is then investigated. Finally, several deep learning models are used to apply ensemble learning and transfer learning to the data given. According to research, transfer learning surpasses domain-specific models. According to the findings, MobileNet and InceptionResnetV2 could obtain training accuracy of around 93%.

Item Type:	Thesis (Masters)
Subjects:	Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science 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:	19 Jan 2023 16:41
Last Modified:	06 Mar 2023 13:43
URI:	https://norma.ncirl.ie/id/eprint/6101

Actions (login required)

View Item