Megha D. Tijare Journal of Engineering Research and Application www.ijera.com ISSN: 2248-9622 Vol. 8, Issue 12 (Part -III) Dec 2018, pp 28-31 www.ijera.com DOI: 10.9790/9622-0812032831 28 | Page Detecting Skin Disease by Accurate Skin Segmentation Using Various Color Spaces Megha D. Tijare*, Dr. V. T. Gaikwad** *(Department of Electronics & Telecommunication Engg.SIPNA COET **(Prof. Department of Electronics & Telecommunication Engg. SIPNA COET Corresponding Author: Megha D. Tijare ABSTRACT Skin diseases which may be of the bacterial, fungal, allergies, enzyme etc. are very harmful for the skin and can spread throughout if not detected accurately as early as possible. So becomes necessary to detect the type disease accurately in early stage and control it by taking proper precautions is demanding now days. So the automatic image analysis method can work really good in this way and is the heart of image processing. Especially in medical field it becomes useful for providing the quantitative information related with the skin disease. So proved as a early warning tool for future problems during the treatment. Now a day's there is a need to perform the detection of disease accurately without any penetration in the body that's why simply digital images of affected skin region are captured by the camera can be processed by using image processing tools. There are many sub techniques of image processing tools are work and played important role in the area of research. This paper presents the detection of skin disease by image processing tools like preprocessing, segmentation, feature extraction, classification and uses specific techniques for the relative steps. While doing the segmentation it uses various color spaces like RGB, Ycbcr, HSV etc. so that accurate skin segmentation will be done for proper feature extraction and classification so the disease can be detected accurately and can improve the efficiency of the system. Also gives the type and percentage of diseases. Keywords - color spaces, feature extraction, Image preprocessing, KNN, segmentation. -------------------------------------------------------------------------------------------------------------------------------------- Date of Submission: 20-12-2018 Date of Acceptance: 04-01-2019 -------------------------------------------------------------------------------------------------------------------------------------- INTRODUCTION Skin is the covering organ of human body. All the organs inside the human body are completely protected by skin. Therefore it is important to give the attention towards the complete care of skin. Because any change in its normal functioning can cause to affect the other parts of body [2]. Many of the skin diseases can be caused by the exposure to the ultraviolet rays come from sun and manmade sources [5]. There are two major layers in human skin such as epidermis and dermis. The outer layer of the skin is called as epidermis consist of three cells such as Basel cells, Squamous cells and melanoma that provide the skin its color and protected from damage. Inner layer is the dermis contains nerves, blood vessels and sweat glands. Any of the wrong functioning of these cells can cause spot on the skin which is infected and called as lesion area. Which is the first notation of skin disease if not detected in early stage can spread throughout [9]. Medical field depends more on computer aided diagnosis because early detection of skin disease is more complex for dermatologist and if detected then it is very time consuming and performed by the penetration in the body as a form of injection. So the simple way is to investigate the digital images are by means of image processing techniques which are popular now a days and becomes a interesting field for the researchers. These techniques took the less time for detection if the relevant data is available .For capturing the data doesn't requires the physical contact with skin [2]. So the proposed methodology is in the same direction, follows the steps as shown in figure. RELATED WORK It gives an idea about latest approaches in the various fields of interest and for finding out the new research based on the existing. Various approaches have been proposed in the direction of improving the efficiency of the system. Various methods are used in such a direction so as to increase the accuracy of the system. Some of the approaches are as follows in the field of digital image processing for detection of skin diseases. 2.1. Martin Kreutz, Malik Anschutz [1] Presents a paper which uses the combination of DIP and ANN for achieving fast and reliable diagnosis of malignant and benign skin lesions coming from DANAOS data collection. RESEARCH ARTICLE OPEN ACCESS