International Journal of Advance Research In Science And Engineering http://www.ijarse.com IJARSE, Vol. No.4, Special Issue (02), February 2015 ISSN-2319-8354(E) 727 | Page SKIN TEXTURE ANALYSIS FOR MEDICAL DIAGNOSIS - A REVIEW Kriti Jain 1 ,Nidhi Sethi 2 ,Vishal Sharma 3 1 Department of Computer Science and Engineering, DIT University, Dehradun, (India) 2,3 Assistant professor in Computer Science Department, DIT University, Dehradun, (India) ABSTRACT As the technicality in the day to day life is increasing, the world is getting highly dependent on different devices for almost each and every work. Every individual is dependent on these devices for accurate and efficient results. Also, the manual work has been reduced to a great extent. Skin texture analysis is one of the major issues in the field of medical diagnosis. Various types of skin diseases are affecting human life. To treat in an efficient and significant manner and to provide the best ailment, the disease has to be diagnosed properly. Hence, the texture of skin is analysed based on various features and characteristics so that the discrepancies can be avoided during the treatment. Therefore, the purpose behind this review paper is to create a base for the research and introduction of the algorithm that is to be used for the same i. e. GLCM and Wavelet Decomposition method. Keywords: Diagnosis,GLCM, Haar Wavelet, Markov Random Field, Skin Diseases. I INTRODUCTION Texture and colour of human skin has proven to be the most important aspect of several imaging systems. Human texture reproduction has been very beneficial in medical diagnosis, cosmetic analysis etc. [1].The basis on which the human is able to discriminate between surfaces and objects is texture analysis. The skin texture has a close relation with the individual’s diet, hormones, hydration and any allergic symptoms.[2] Various properties correlate with the skin texture e.g. skin dryness, fungus, allergic symptoms etc. Significant amount of work has been performed on analysis of the skin texture till date. Previously the image is analysed for extracting haemoglobin and melanin components by independent component analysis. The image has been synthesized for the comparison. [3] The skin colour image is also decomposed by YuantingGu and Enhua Wu to the four texturecomponents by multi-resolution analysis using wavelettransform for synthesizing the image and separating it from the original image. [4] Texture contents are also decomposed for separating the image of skin into the basic shape and the feature vector based on which the texton (pixel) change is observed to determine the pattern and structure. This process is done under image synthesis.