Gray Image Colorization Based on General Singular Value Decomposition and YCbCr Color Space
Keywords:
Image processing, colorization, GSVD, YCbCr color space, gray imageAbstract
Colorization the gray image is the process of adding colors to the gray image without prior knowledge about the real colors of image. In general, the term and process of colorization is an active area of many research, and challenging for many researchers. In this paper we suggest new method for automatic colorizing gray image depending on using generalized singular value decomposition (GSVD) algorithm with YCbCr color space. Up to our knowledge this the first work uses the GSVD algorithm in this field. We suggested to use reference color image. Both reference image and gray image transformed to YCbCr color space (gray image converted to 3D image by redundant the gray image three times). GSVD applied for each block from gray image with all blocks from reference image (converted to gray image) to find the best blocks can be combines together from both reference image and gray image. The (Y, Cb, Cr) from YCbCr color space of gray and reference images combines together (Y channel corresponding to block of gray image combine with (Cb and Cr) channels from block corresponding to reference image). This process continues for all blocks of gray image. Finally, the resulted image (YCbCr image) transform to RGB image (colorized image). The results were promised and dependable.References
Charpiat, G., Hofmann, M. & Schölkopf, B. (2008) Automatic Image Colorization via Multimodal Predictions, European Conference on Computer Vision (ECCV) 2008, pp 126-139.
Bisht, U. & Patnaik, T. (2015) Overview of Automatic Image Colorization Schemes, International Journal of Advanced Engineering and Global Technology, Vol-03, Issue-10, pp 1283-1287, November.
Kumar, S. & Swarnkar, A. (2012) Gray image colorization in YCbCr color space. Emerging Technology Trends in Electronics, Communication and Networking (ET2ECN), 1st International Conference on IEEE.
Wang, Huanjuan, & et al (2012) Novel colorization method based on correlation neighborhood similarity pixels’ priori." Signal Processing (ICSP), IEEE 11th International Conference on. Vol. 2. IEEE.
Devi, Sarada, M. & Mandowara, A. (2012) Extended performance comparison of pixel window size for colorization of grayscale images using YUV color space. Engineering (NUiCONE), Nirma University International Conference on IEEE.
S. D., Garg, Ghewade, Jagdale, Mahajan & et al. (2015). Performance assessment of assorted similarity measures in gray image colorization using LBG vector quantization algorithm, International Conference on Industrial Instrumentation and Control (ICIC), IEEE.
Hu, Min, Bo Ou, and Yi Xiao (2017) efficient image colorization based on seed pixel selection. Multimedia Tools and Applications76.22.
Deshpande, Aditya, Rock, J. & Forsyth, D. (2015) Learning large-scale automatic image colorization. Proceedings of the IEEE International Conference on Computer Vision.
Okura, Fumio & et al. (2015) Unifying color and texture transfer for predictive appearance manipulation. Computer Graphics Forum. Vol. 34. No. 4.
Li, Bo, Lai, Y. & Paul L. (2017) Rosin. Example-based image colorization via automatic feature selection and fusion. Neurocomputing 266: 687-698.
Kerr & Douglas A. (2005) Chrominance subsampling in digital images. The Pumpkin, (1), November.
Abdi & Hervé (2007) Singular value decomposition (SVD) and generalized singular value decomposition. Encyclopedia of measurement and statistics (2007): 907-912.
Wei, Yimin, Xie, P. & Zhang, L. (2016) Tikhonov regularization and randomized GSVD. SIAM Journal on Matrix Analysis and Applications 37.2 (2016): 649-675.
Loan, V. & Charles (1985) Computing the CS and the generalized singular value decompositions. Numerische Mathematic 46.4 (1985): 479-491.
Hansen & Christian, P. (1990) Relations between SVD and GSVD of discrete regularization problems in standard and general form. Linear Algebra and Its Applications 141: 165-176.
