Influence of Ambiguity Cluster on Quality Improvement in Image Compression

Authors: Safaa Al-Ali, Ahmad Shahin, Fadi Chakik

Abstract:

Image coding based on clustering provides immediate access to targeted features of interest in a high quality decoded image. This approach is useful for intelligent devices, as well as for multimedia content-based description standards. The result of image clustering cannot be precise in some positions especially on pixels with edge information which produce ambiguity among the clusters. Even with a good enhancement operator based on PDE, the quality of the decoded image will highly depend on the clustering process. In this paper, we introduce an ambiguity cluster in image coding to represent pixels with vagueness properties. The presence of such cluster allows preserving some details inherent to edges as well for uncertain pixels. It will also be very useful during the decoding phase in which an anisotropic diffusion operator, such as Perona-Malik, enhances the quality of the restored image. This work also offers a comparative study to demonstrate the effectiveness of a fuzzy clustering technique in detecting the ambiguity cluster without losing lot of the essential image information. Several experiments have been carried out to demonstrate the usefulness of ambiguity concept in image compression. The coding results and the performance of the proposed algorithms are discussed in terms of the peak signal-tonoise ratio and the quantity of ambiguous pixels.

Keywords: Ambiguity Cluster, Anisotropic Diffusion, Fuzzy Clustering, Image Compression.

Digital Object Identifier (DOI): doi.org/10.5281/zenodo.1073447

References:

[1] G. K. Wallace. "The JPEG Still-Picture Compression Standard". Communications of ACM, pp.30-44, Apr. 1991.
[2] N. Somasundaram and Y. V. RamanaRao." Modified LOG-EXP Based Image Compression Algorithm", IJCSNS International Journal of Computer Science and Network Security, VOL.8 No.9, 179-184, Sept. 2008.
[3] A. Gersho and R. M. Gray.,"Vector Quantization and Signal Compression", Kluwer Academic, 1991.
[4] J. Weickert, "Anisotropic Diffusion in Image Processing", B.G. Teubner Stuttgart 1998.
[5] Martin Welk, David Theis, Thomas Brox, and Joachim Weickert, "PDEBased Deconvolution with Forward-Backward Diffusivities and Diffusion Tensors", in Scale Space 2005, Springer LNCS 3459, pp. 585- 597, Hofgeismar, Germany, Apr. 2005
[6] A.Shahin, F.Chakik, and S.Al-Ali, "Complexity Reduction and Quality Enhancement in Image Coding", ICIMT 2012, Kuala Lumpur, Malaysia, December 2012, submitted for publication.
[7] Lizarazo, Ivan and Barros," Fuzzy image segmentation for urban landcover classification", Photogrammetric Engineering and Remote Sensing, Joana, pp. 151-162, 2010.
[8] J. C. Bezdek. "Pattern Recognition with Fuzzy Objective Function Algoritms", Plenum Press, New York, 1981.
[9] Miin-Shen Yang, Yu-Jen HU, Karen Chia-Ren-Lin, Charles Chia-Ren- Lin, "Segmentation techniques for tissue differentiation in MRI of Ophthalmology using fuzzy clustering algorithms", Magnetic Resonance Imaging (20),pp.173-179, ELSEVIER, 2002
[10] P. Perona and J. Malik, "Scale-space and edge detection using anisotropic diffusion," IEEE Trans. Pattern Anal. Machine Intell., vol.12, pp. 629-639, 1990.
[11] M. MÉNARD. "The fuzzy c+2 means: solving the extended ambiguity reject in clustering". Traitement du Signal. Volume 16 - n°2. 1999.