Dr. H. B. Kekre,Dr. Sudeep D. Thepade,Dr. Tanuja K. Sarode,Ms. Nikita Bhandari/International Journal of Engineering Research and Applications (IJERA) ISSN: 2248-9622 www.ijera.com Vol. 1, Issue 4, pp.1274-1283 1274 | P a g e Colorization of Grayscale Images using LBG VQ Codebook for different Color Spaces Dr. H. B. Kekre 1 , Dr. Sudeep D. Thepade 2 , Dr. Tanuja K. Sarode 3 , Ms. Nikita Bhandari 4 1, 2, 4 (Department of Computer Engineering, Mukesh Patel School of Technology Management & Engineering, NMIMS University, Vile-parle (w), Mumbai-50.) 3 (Department of Computer Engineering , Thadomal Shahani Engineering College, Mumbai University, Bandra(W), Mumbai-50.) ABSTRACT The paper presents an innovative technique for colorization of grayscale images. Here the colors from some source color image are picked up and sprayed into the to be colored grayscale image. The color palette used in colorization technique discussed here is generated using the Linde Buzo and Gray (LBG) codebook. The technique is tested using various VQ codebook sizes like 32, 64, 128, 256 and 512. Also various color spaces like RGB, Kekre’s LUV, YCbCr, YUV, YIQ, and Kekre’s Biorthogonal color spaces are considered for colorization of grayscale images. In all 5 VQ codebook sizes and 8 color spaces give total 40 versions of proposed colorization method. All proposed versions are implemented and tested for colorization of grayscale images from different classes. Keywords - About five key words in alphabetical order, separated by comma I. INTRODUCTION Colorization is a computer assisted process of adding color to a monochrome image or movie. The task of color traits transfer to the gray scale image is called as colorization of gray scale image. It is difficult because it involves three dimensional RGB pixel values to an image which varies along one dimension (luminance or intensity)[1,24,25]. Since different colors may have the same luminance value but vary in hue or saturation, the problem of colorizing grayscale images has no inherently unique solution [2]. Due to these ambiguities, human interaction usually plays a large role in the colorization process. Even in the case of pseudocoloring, [3, 4] where the mapping of luminance values to color values is automatic, the choice of the color- map is commonly determined by human decision. Pratt [4] describes this method as an image enhancement technique because it can be used to enhance the detect ability of detail within the image [1]. Grayscale image colorization can find its applications in black and white photo editing [1,5], classic movies colorization [3,6,7] and scientific illustrations [1,2]. Colorization can increase dramatically the visual appeal of grayscale images and perceptually enhance scientific illustrations. In medicine [1,4] , image modalities which only acquire grayscale images such Magnetic Resonance Imaging (MRI), X-ray and Computerized Tomography (CT) images can be enhanced with color for presentations and demonstrations. Pseudocoloring is a common technique for adding color to grayscale images such as X-ray, MRI, scanning electron microscopy (SEM) and other imaging modalities in which color information does not exist. In this paper we present novel colorization method using LGB Vector Quantization codebook, five different codebook sizes varying from 32 to 512 are used to generate color pallet and eight different color spaces are used leading to 40 different colorization methods. II. COLOUR SPACE USED FOR EXPERIMENTATION This section presents various color spaces as Kekre‟s LUV, YCbCr, YUV, YIQ and Kekre‟s Biorthogonal color spaces. 2.1 Kekre‟s LUV Color Space [8, 9] Here we have used Kekre‟s LUV color Space. Where L gives luminance and U and V gives chromaticity values of color image. Positive value of U indicates prominence of red component in color image and negative value of V indicates prominence of green component. This needs the conversion of RGB to LUV components. The RGB to LUV conversion matrix given in equation 1 gives the L, U, V components of color image for respective R, G, B components.