• Siddharth Kumar

Over the last decades, digital image processing based skin lesion segmentation and classification have been improving steadily to provide a more accurate classification results in the area of medical science. Segmentation and classification of the skin lesion from the non dermoscopic images is very challenging task due to the complex structural properties of the images and need improvisation in the existing work by utilization of improvement in segmentation and feature selection approach to select on optimal feature according to the nevus and melanoma type of skin cancer. A research based on the combination of various algorithms for nevus and melanoma classification has been presented in this paper. In this research, we develop an improved K-means based Nevus and Melanoma Classification (NMC) system from non-dermoscopy images using Deep Learning is proposed. The concept of Speed up Robust Feature (SURF) along with the hybridization of Artificial Bee colony (ABC) and Cuckoo Search Algorithm (CSA) for feature selection being used. Here, Convolutional Neural Network (CNN) is used as an Artificial Intelligence (AI) technique with that helps to train the NMC system. By utilizing the concept of improved K-means, the accuracy of skin lesion segmentation is increases in terms of accuracy. So, classification efficiency of the NMC model is also improved as compare to existing works. At last, the performance of the NMC system is calculated to validate the model and this shows that it is possible to use optimized SURF as a feature extraction technique in order to classify the nevus and melanoma skin cancer with minimum error rate and the simulation results clearly show the effectiveness of proposed NMC system.

Nevus and Melanoma Classification (NMC) System, SURF Descriptor, ABC, CSA, Artificial Intelligence (AI) and CNN

"An Improved K-means based Nevus and Melanoma Classification System from Non-Dermoscopy Images using Deep Learning", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.9, Issue 7, page no.c139-c151, July-2022, Available :http://www.jetir.org/papers/JETIR2207216.pdf

"An Improved K-means based Nevus and Melanoma Classification System from Non-Dermoscopy Images using Deep Learning", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.9, Issue 7, page no. ppc139-c151, July-2022, Available at : http://www.jetir.org/papers/JETIR2207216.pdf