Adaptive Fine-Tuned AdaBoost and Improved Firefly Algorithm for Skin Cancer Detection
Anupama Damarla, D. Sumathi · Traitement du signal · 2024
Skin cancer is a major malignancy caused by exposure to the sun's ultraviolet radiation.The patients are completely oblivious to the early stages of skin cancer development.Computer Vision Systems (CVS) that evaluate digital images of skin lesions are being used in research to accomplish an early diagnosis of melanoma.These methods give an automated analytical model for a precise and quick assessment of the lesions.In this study, we propose a Median Filter (MF) and Contour-Based Image Enhancement (CIE) for pre-processing, Inception v3 Clustering Algorithm (IV3-CA) for data segmentation, Inception ResNet v2 (IRV2) approach for feature extraction.Furthermore, the efficiency of the CNN was enhanced using an Improved Firefly Algorithm (IFFA) and classified with the Adaptive Fine-Tuned AdaBoost algorithm.The performance is investigated on the ISIC-2017 dataset using 2000 images.According to assessments, the modified model has remarkable identification benefits and has achieved an accuracy of 97.14 percent.The results show that the suggested method performs better than the current approach.