Fuzzy clustering for colour reduction in images

The aim of colour quantisation is to reduce the number of distinct colour in images while preserving a high colour fidelity as compared to the original images. The choice of a good colour palette is crucial as it directly determines the quality of the resulting image. Colour quantisation can also be seen as a clustering problem where the task is to identify those clusters that best represent the colours in an image. In this paper we investigate the performance of various fuzzy c-means clustering algorithms for colour quantisation of images. In particular, we use conventional fuzzy c-means as well as some more efficient variants thereof, namely fast fuzzy c-means with random sampling, fast generalised fuzzy c-means, and a recently introduced anisotropic mean shift based fuzzy c-means algorithm. Experimental results show that fuzzy c-means performs significantly better than other, purpose built colour quantisation algorithms, and also confirm that the fast fuzzy clustering algorithms provide similar quantisation results to the full conventional fuzzy c-means approach.

[1]  Daoqiang Zhang,et al.  Robust image segmentation using FCM with spatial constraints based on new kernel-induced distance measure , 2004, IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics).

[2]  John F. Kolen,et al.  Reducing the time complexity of the fuzzy c-means algorithm , 2002, IEEE Trans. Fuzzy Syst..

[3]  S.M. Szilagyi,et al.  MR brain image segmentation using an enhanced fuzzy C-means algorithm , 2003, Proceedings of the 25th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (IEEE Cat. No.03CH37439).

[4]  Brian A. Wandell,et al.  Color image fidelity metrics evaluated using image distortion maps , 1998, Signal Process..

[5]  Gerald Schaefer,et al.  Retrieval of palettized color images , 1999, Electronic Imaging.

[6]  Lawrence O. Hall,et al.  Fast Accurate Fuzzy Clustering through Data Reduction , 2003 .

[7]  Dorin Comaniciu,et al.  Mean shift analysis and applications , 1999, Proceedings of the Seventh IEEE International Conference on Computer Vision.

[8]  Richard J. Hathaway,et al.  On efficiency of optimization in fuzzy c-Means , 2002, Neural Parallel Sci. Comput..

[9]  Paul S. Heckbert.  Color image quantization for frame buffer display , 1998 .

[10]  Daoqiang Zhang,et al.  Fast and robust fuzzy c-means clustering algorithms incorporating local information for image segmentation , 2007, Pattern Recognit..

[11]  Dorin Comaniciu,et al.  Mean Shift: A Robust Approach Toward Feature Space Analysis , 2002, IEEE Trans. Pattern Anal. Mach. Intell..

[12]  Bo Thiesson,et al.  Image and Video Segmentation by Anisotropic Kernel Mean Shift , 2004, ECCV.

[13]  Aly A. Farag,et al.  A modified fuzzy c-means algorithm for bias field estimation and segmentation of MRI data , 2002, IEEE Transactions on Medical Imaging.

[14]  Paul Scheunders,et al.  A genetic c-Means clustering algorithm applied to color image quantization , 1997, Pattern Recognit..

[15]  James C. Bezdek,et al.  A Convergence Theorem for the Fuzzy ISODATA Clustering Algorithms , 1980, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[16]  G. Schaefer,et al.  A mean shift based fuzzy c-means algorithm for image segmentation , 2008, 2008 30th Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[17]  Tzong-Jer Chen,et al.  Fuzzy c-means clustering with spatial information for image segmentation , 2006, Comput. Medical Imaging Graph..

[18]  Jacek M. Leski,et al.  Towards a robust fuzzy clustering , 2003, Fuzzy Sets Syst..

[19]  Lawrence O. Hall,et al.  Fast fuzzy clustering , 1998, Fuzzy Sets Syst..

[20]  Gerald Schaefer,et al.  Colour map design through optimization , 2007 .

[21]  Michael Gervautz,et al.  A simple method for color quantization: octree quantization , 1990 .

[22]  Anthony H. Dekker,et al.  Kohonen neural networks for optimal colour quantization , 1994 .