Hybrid Machine Learning Approach for Skin Disease Detection Using Optimal Support Vector Machine
Dermoscopy is the one of the major discussing topic in medical field to detect the skin diseases. Due to the bad frequencies, low contrast and noises the positive result of the dermoscopy is unpredictable. In this paper, a hybrid machine learning model of Singular Value Decomposition (SVM) based Whale Optimization Algorithm (WOA) is proposed for the identification of skin disease. Initially, the image database is segmented using level set approach. In order to retrieve the feature vectors from the segmented image, extraction of features from those datasets are carried out. Hence, the feature vectors are extracted using histogram and Local binary pattern (LBP) method. Once feature is extracted, the skin disease classification is done using Whale optimization (WOA) based Support Vector Machine (SVM). The simulation results show that the proposed method outperforms many other algorithms in terms of accuracy and other performance measures in the identification of skin disease.
KeywordsDermoscopy Skin image Segmentation Classifier WOA-SVM
- 1.Yadav, N., Narang, V.K., Shrivastava, U.: Skin diseases detection models using image processing. Int. J. Comput. Appl. 137(12), 0034–0039 (2016)Google Scholar
- 2.Barata, C., Marques, J.S.: A bag of features approach for the classification of melanomas in dermoscopy images. In: Computer Vision Techniques for the Diagnosis of Skin Cancer, pp. 49–69. Springer (2014)Google Scholar
- 3.Choudhari, S., Biday, S.: Artificial neural network for skin cancer detection. Int. Emerg. Trends Technol. Comput. Sci. 3(5), 147–153 (2014)Google Scholar