On spatial quantization of color images
Image quantization and dithering are fundamental image processing problems in computer vision and graphics. Both steps are generally performed sequentially and, in most cases, independent of each other. Color quantization with a pixel-wise defined distortion measure and the dithering process with its local neighborhood typically optimize different quality criteria or, frequently, follow a heuristic approach without reference to any quality measure.
In this paper we propose a new model to simultaneously quantize and dither color images. The method is based on a rigorous cost-function approach which optimizes a quality criterion derived from a simplified model of human perception. Optimizations are performed by an efficient multiscale procedure which substantially alleviates the computational burden.
The quality criterion and the optimization algorithms are evaluated on a representative set of artificial and real-world images thereby showing a significant image quality improvement over standard color reduction approaches.
KeywordsCost Function Color Space Color Quantization Error Diffusion Iterative Conditional Mode
Unable to display preview. Download preview PDF.
- 1.P. Heckbert, “Color image quantization for frame buffer displays,” Computer Graphics, vol. 16, no. 3, pp. 297–307, 1982.Google Scholar
- 2.G. Braudaway, “A procedure for optimum choice of a small number of colors from a large palette for color imaging,” in Electron Imaging'87, 1987.Google Scholar
- 4.A. Jain and R. Dubes, Algorithms for Clustering Data. Prentice Hall, 1988.Google Scholar
- 6.R. Floyd and L. Steinberg, “An adaptive algorithm for spatial greyscale,” in Proc. SID, Vol. 17, No. 2, Wiley, 1976.Google Scholar
- 7.L. Akarůn, D. özdemir, and ö. Yalcun, “Joint quantization and dithering of color images,” in Proceedings of the International Conference on Image Processing (ICIP'96), pp. 557–560, 1996.Google Scholar
- 9.T. Hofmann, J. Puzicha, and J. Buhmann, “Deterministic annealing for unsupervised texture segmentation,” in Proc. of the EMMCVPR'97, LNCS 1223, pp. 213–228, 1997.Google Scholar
- 11.J. Puzicha and J. Buhmann, “Multiscale annealing for real-time unsupervised texture segmentation,” Tech. Rep. IAI-97-4, Institut für Informatik III (a short version appeared in: Proc. ICCV'98, pp. 267–273), 1997.Google Scholar
- 12.C. I. de L'Eclairage, “Colorimetry.” CIE Pub. 15.2 2nd ed., 1986.Google Scholar
- 13.D. Alman, “Industrial color difference evaluation,” Color Res. Appl. 18 137–139, 1993.Google Scholar
- 14.G. Bilbro, W. Snyder, S. Garnier, and J. Gault, “Mean field annealing: A formalism for constructing GNC-like algorithms,” IEEE Transactions on Neural Networks, vol. 3, no. 1, 1992.Google Scholar
- 15.T. Hofmann, J. Puzicha, and J. Buhmann, “A deterministic annealing framework for textured image segmentation,” Tech. Rep. IAI-TR-96-2, Institut für Informatik III, 1996.Google Scholar
- 16.J. Puzicha, M. Held, J. Ketterer, J. Buhmann, and D. Fellner, “On spatial quantization of color images,” Tech. Rep. IAI-TR-98-1, Department for Computer Science III, University Bonn, 1998.Google Scholar
- 17.M. Gervauz and W. Purgathofer, “A simple method for color quantization: Octree quantization,” in Graphic Gems, pp. 287–293, Academic Press, New York, 1990.Google Scholar