NORMA eResearch @NCI Library

Classification of Melanoma using Transfer Learning and Deep Learning Neural Networks

Goyal, Vishal Satishbhai (2020) Classification of Melanoma using Transfer Learning and Deep Learning Neural Networks. Masters thesis, Dublin, National College of Ireland.

[thumbnail of Master of Science]
Preview
PDF (Master of Science)
Download (10MB) | Preview
[thumbnail of Configuration manual]
Preview
PDF (Configuration manual)
Download (16MB) | Preview

Abstract

The world is suffering from many fatal skin disease, one such disease causing fatalities is Melanoma skin cancer which is being researched from the past many years. Different Deep learning Neural Networks have achieved optimistic results in identifying the Melanoma. However, due to the similar optics of malignant and benign tumor images, it becomes very difficult to get precise results and so a huge improvement is needed. This project helps the automatic classification of melanoma images using deep learning and transfer learning models. Image Augmentation and pre-processing are performed on the images for the model to achieve better accuracy and at last, all the applied models are evaluated based on accuracy and loss. After comparison, it was observed that MobileNet, LeNet, and ResNet achieved better accuracy as compared to CNN and AlexNet. Highest accuracy of 75.38% was achieved by MobileNet.
Area: Deep Neural Network, AlexNet, LeNet, ResNet-50, MobileNet, CNN, Image Processing, BlackHat Filter, Adam Optimization.

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: Dan English
Date Deposited: 22 Jan 2021 12:51
Last Modified: 22 Jan 2021 12:51
URI: https://norma.ncirl.ie/id/eprint/4444

Actions (login required)

View Item View Item