In this paper, a color image segmentation approach based on holographic thresholding and region merging is presented. The holographic image considers both the occurrence of the gray levels and the neighboring nearest value among-st pixels. Thus, it employs both the local and global information. Holographic analysis is performed using fuzzy entropy as a tool for finding all major kindred regions at the first stage followed by region merging process which is carried out based on color similarity among-st these regions to avoid over segmentation. The proposed Kindred -based approach (KOB) is compared with the histogram-based approach (HIB). The experimental results demonstrate that the KOB can find similar regions more effectively than HIB does, and thus can solve the problem of discriminating shading in color images to a greater extent.

Color Image Segmentation, Fuzzy Logic, Region Merge, Color Space, Thresholding

[1]. H. D. Cheng, Ying Sun, X. H. Jiang and Jingli Wang. �Color Image Segmentation: Advances and Prospects.�, Pattern Recognition, 2001.
[2]. C. K. Yuang and W. H. Thsai. �Reduction of Color Space Dimensionality by Moment- preserving Thresholding and Its Application for Edge Detection in Color Images.� Pattern Recognition Letters, Vol.17, 485-490, 1996.
[3]. Y. Ohta. T. Khanade and Sakai T. �Color Information for Region Segmentation.� Computer Graphics and Image Processing. Vol.13, 220-241,1980.
[4]. D. Tseng and C. H. Chang.�Color Segmentation Using Perceptual Attributes.� IEEE International Conference on Image �Processing-A,226-231,1992.
[5]. R. Ohlander, K. Price and Reddy D.R. �Picture Segmentation Using A Recursive Region Splitting Method.� Computer Graphics and Image Processing. 8, 310-333,1978.
[6]. N.Ito. et al., �The Combination of Edge Detection and Region Extraction in Non- parametric Color Image Segmentation� Information Sciences,92, 280-294,1996.
[7]. R. I. Taylor and Lewis P.H., �Color Image Segmentation Using Boundary Relaxation,� IEEE International Conference on Image Processing-C, 720-724,1992.
[8]. R. Schettini, �Segmentation Algorithm for Color Images,� Pattern Recognition Letter,14, 500-506,1993.
[9]. R. M. Haralick and L.G. Shapiro, � An Image Segmentation Technique.� Computer Vision, Graphics and Image Processing,29,120-132,1985.
[10]. P. K. Sahoo, et, al., �A Survey of Thresholding Techniques for image processing,� Computer Vision, Graphics and Image Processing, 41,255-260,1988.
[11]. L. Spirkovski. �A Summary of Image Segmentation Techniques� NASA Technical Memorandum 104022, June 1993.
[12]. C. H. Chen, �On the Statistical Image Segmentation Techniques,� Proceedings, IEEE Conference on Pattern Recognition and Image Processing, 260-266,1981.
[13]. P. W. M. Tsang and W. H. Tsang, �Edge Detection techniques on Object Color,� IEEE International Conference on Image Processing,C, 1040-1052,1996.
[14]. M. T. Orchard and C. A. Bouman, �Color Quantization of Images,� IEEE Transactions on Signal Processing, Vol.39, No. 12, 2675-2690, December 1991.
[15]. M. Pietikainen, et al., �Accurate Color Discrimination with Classification Based on Feature Distributions,� International Conference on Image Processing,C,830-838,1998.
[16]. T. L. Huntsberger, C. L. Jacobs and R. L. Cannon, �Iterative Fuzzy Image Segmentation,� Pattern Recognition,Vol. 18, No. 2,130-140, 1985.
[17]. T. Carron and P. Lambert, �Color Edge Detector Using Jointly Hue, Saturation and Intensity,� International Conference on Image Processing , Austin, USA, 975-985,1994.
[18]. M. Chapron, �A new Chromatic Edge Detector Used for Color Image Segmentation,� IEEE International Conference on Image Processing,A, 310-315,1992.
[19]. J. Kender, �Saturation, Hue, and Normalized Color Calculation, Digitization Effects and Use,� Computer Science Technical Report, Carnegie Mellon University,1976.