Analysis of Dermoscopic Images using Multiresolution Approach
K.S Rajasekhar1, T. Ranga Babu2
1K.S Rajasekhar, Research Scholar, Department of ECE, Acharya Nagarjuna University, Guntur (A.P), India.
2Dr. T.Ranga Babu, Professor, Department of ECE, RVR &JC College of Engineering & Technology, Guntur (A.P), India.
Manuscript received on 05 February 2019 | Revised Manuscript received on 13 February 2019 | Manuscript published on 28 February 2019 | PP: 41-48 | Volume-8 Issue-4, February 2019 | Retrieval Number: D2617028419/19©BEIESP
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Abnormal growth of cells in any part of the body is called cancer. Cancer that is formed on skin is called skin cancer. Life span of a cancer patient can be increased by the early detection of tumor part. This paper deals with classification of dermoscopic images, i.e. benign or malignant based on coefficients extracted from multiresolution analysis based wavelet functions and tetrolet transform. Statistical texture features such as Mean, Standard Deviation, Kurtosis and Skewness are calculated from the coefficients of the multiresolution transfroms. The Gray Level Co-occurence Matrix(GLCM) is calculated for the dermoscopic images from which features such as homogenity, energy and entropy are calculated. In addition to these shape features are also taken into consideration. K-Nearest Neighbor(KNN) classifier is used for classification of dermoscopic images. In this work, dermoscopic images are obtained from the International Skin Imaging Archive (ISIC). The performance of the system is evaluated using accu-racy, sensitivity and specificity. The area under the curve(AUC) demonstrates the superiority of tetrolet transform.
Keyword: Dermoscopic Images, Texture Features, GLCM Features, Shape Features, KNN Classifier, Accuracy, Sensitivity, Specificity And AUC.
Scope of the Article: Image analysis and Processing