IJRET: International Journal of Research in Engineering and Technology eISSN: 2319-1163 | pISSN: 2321-7308 _______________________________________________________________________________________ Volume: 04 Issue: 05 | May-2015, Available @ http://www.ijret.org 76 FACE SKIN COLOR BASED RECOGNITION USING LOCAL SPECTRAL AND GRAY SCALE FEATURES T.Anil Raju 1 , B.Ajanta Reddy 2 , N. Adinarayana 3 1 Asosciate professer, Lakireddy Bali Reddy college of Engineering, Mylavaram-521230, Krishna (Dt) 2 Student of M.Tech, Lakireddy Bali Reddy college of Engineering, Mylavaram-521230, Krishna (Dt) 3 Student of M.Tech, Lakireddy Bali Reddy college of Engineering, Mylavaram-521230, Krishna (Dt) Abstract Human face conveys more information about identity of person. Human face recognition is one of the most challenging problem and it can be used in many applications at different security places in airports, defense and banking sectors etc.In this work used colored features obtained from color segmentation because in real time scenario color provides the more information than gray scale image but it has a drawback. To overcome this drawback gray scale feature extracted from co-occurrence matrix of an image and for efficient face recognition of human Face under different illumination conditions spectral features can be extracted from face texture. These three feature vectors concatenated into a single feature vector and applied Lenc-Kral matching technique to measure similarity between the database and query image, the similarity is high then face is recognized. Keywords: Face recognition, illumination condition, local texture features, color segmentation. --------------------------------------------------------------------***------------------------------------------------------------------ 1. INTRODUCTION Face recognition has more robust technology compare to other biometrics [1] because advantage of face recognition is it does not need any physical interaction between person and device. The main idea of face recognition is human face, because it conveys more information and might be unique features like texture, eyes, nose and mouth etc. among these, texture features are more important [2] the basic idea about texture features is face. It has a lot of applications like Airport, defense, banking sectors and many historical places etc. The human face of a different skin colored people have different skin color due to ultraviolet radiation penetration because skin is more exposure to the UV radiation levels [3]. Skin reflectance is strongly correlated with absolute latitude and UV radiation levels and texture and shape based on their surroundings. Different approaches has been proposed for face recognition, including Principal component analysis, [4] Linear discriminant analysis [5], independent component analysis [6], kernel methods, neural networks, elastic bunch graph method and wavelets etc.most of the methods were developed on well-controlled environments like uniform background and well aligned faces, but it is impossible in uncontrolled environment [7] to recognize face due to illumination, facial expressions and aging etc. To overcome these problems in this proposed work consists of three categories for efficient face recognition under different lighting conditions, those are 1.color segmentation 2.Extraction of feature vectors 3.vector comparision.In color segmentation YCbCr color space model was used for extracted the colored features from skin colored human faces . In Extraction of feature vectors, extracted the two types of features those are local texture features and gay level features. Local texture features are extracted from face texture because face has a composition of micro patterns is called as texture pattern. The main purpose of these features is to recognizing the human face under different lighting conditions. Gray level features extracted from co-occurrence matrix of image for efficient face recognition. In vector comparison, combined these features into a single vector and applied Lenc-kral matching technique. In this technique feature vectors of database and query image compared and measure similarity between these images, similarity is high then face is recognized. 2. COLOR SEGMENTATION Color provides much information about the image than that of gray image.Numorous methods proposed for color segmentation like, RGB, normalized RGB, CMY, CMYK, but these are not well suited for describing colors in variant illumination environment, so we adopt YCbCr color space. In YCbCr by decoupling the color information into intensity and chromatic components, YCbCr color space omits the intensity components and use only descriptor components for skin detection which can provide robustness against changing intensity [8]. YCbCr is not an absolute color space rather it is way of encoding, where Y is luminance, Cb is the blue chrominance component, Cr is the red chrominance component and it has certain range [9]. RGB-YCbCr Equations: The basic equations to convert between 8-bit digital RGB data and YCbCr are Y=0.299R+0.587G+0.114B (1)