Automated Melanoma Detection Using Sparse Classifier and Feature Selection Based T’test Method
Sophia Jone .J
Assistant Professor, Department of ECE, Bethlahem Institute of Engineering, Tamilnadu,India , IN
Jasmineshenika. J
Assistant Professor, Department of ECE, Bethlahem Institute of Engineering, Tamilnadu,India , IN
Keywords: Cold challenge, Texture features, Principal component analysis, localization of abnormality, skin cancer, Texture, Malignant
Abstract
Melanoma is also referred as malignant melanoma, which is a type of skin cancer and it is developed from the pigment-containing cells which are known as melanocytes. High-level features are needed in the existing method because low classification accuracy occurs. In order to improve existing methods sparse representation method is used. Image analysis tasks are carried out by sparse representation classification and is a pixel-wise classification of images. Image pixels can be classified by sparse classifiers. High classification accuracy can be obtained using the sparse method.T'test method is used when the variances of two normal distributions are unknown and use a small sample size. It examines with two samples whether it is different or not. The results show the accuracy of the proposed recognition system and high sensitivity in melanoma recognition using sparse classifiers and T'test method
How to Cite
References
NA