Downloads: 1 | Views: 39 | Weekly Hits: ⮙1 | Monthly Hits: ⮙1
Research Paper | Computer Science and Information Technology | India | Volume 13 Issue 1, January 2024
A DSegNet Model for the Classification and Detection of Skin Cancer Diseases Using Transfer Learning
Vijay Arumugam R [2] | Saravanan S. [5]
Abstract: Detecting and categorizing diseases in skin, particularly skin cancer, involves applying techniques such as Computer Vision (CV) and Machine Learning (ML) to identify and classify skin ailments. The integration of these methodologies has the potential to assist healthcare professionals in promptly identifying and managing skin conditions, contributing to early diagnosis and effective treatment. The classification and detection of skin cancer using Deep Learning (DL) have proven to be a successful approach for automated disease detection. As a subset of Artificial Intelligence (AI), DL focuses on training neural networks with intricate layers to learn complex representations and patterns from data autonomously. In this research, the Deep-SegNet model is employed to develop the DSegNet method for the Automatic Recognition and Classification of Skin Cancer. The DSegNet approach incorporates multiple stages to enhance accuracy and diagnostic performance. Initially, a pre-processing stage is implemented, involving image resizing and the application of a Bilateral Filter (BF) to improve image quality. Subsequently, SegNet-based segmentation is utilized to identify the disease-affected areas, and feature extraction is performed using the MobileNetV3 architecture. Finally, the extracted features are input into a VGG-19 classification model to differentiate between various types of skin cancer. A thorough analysis of experimental results demonstrates that the DSegNet technique outperforms other recent approaches in terms of performance.
Keywords: Deep Learning, Segmentation, Skin Disease, Machine Learning, Transfer Learning, Computer Vision
Edition: Volume 13 Issue 1, January 2024,
Pages: 1321 - 1326