A NOVAL APPROACH TO DETECT AND CLASSIFY SKIN DISEASE BY ANALYSING MRI IMAGES USING FUZZY SUPPORT VECTOR MACHINE (FSVM)

Authors

  • Arshdeep Kaur1, Harjinder Kaur2 SSIET, Dinanagar

Abstract

Skin lesion is the part of the skin having abnormal growth. The part of skin appearing different then other parts of the skin is generally falls into the category of skin lesion. The proposed work investigate the application of two algorithms to form hybrid approach, Support vector machine(SVM) and Fuzzy Filtering mechanism to detect skin lesion from the images presented. The literature uses coloured image set derived from the internet. The proposed work begins by pre processing of image i.e removing unnecessary artifacts such as hair, shading affects etc. Boundary of image is identified through the next step known as segmentation. In the next step fuzzy filtering is used to identify the features and classification is performed. The simulation is conducted in MATLAB using the application of image processing toolbox. The results obtained are compared with existing segmentation approaches to identify accuracy of result in terms of skin lesion detection. Keywords: Skin Lesion, Pre-processing, Segmentation, SVM, Fuzzy Filtering

Downloads

Published

2017-10-29

How to Cite

Harjinder Kaur2, A. K. (2017). A NOVAL APPROACH TO DETECT AND CLASSIFY SKIN DISEASE BY ANALYSING MRI IMAGES USING FUZZY SUPPORT VECTOR MACHINE (FSVM). International Journal of Engineering Technology and Computer Research, 5(5). Retrieved from http://www.ijetcr.org/index.php/ijetcr/article/view/448

Issue

Section

Articles