Image segmentation is to classify or cluster an image into several regions according to the feature of image, for example, the pixel intensity or the distance measure. Up to now, lots of image segmentation algorithms exist and be extensively applied in science and daily life. [2,3]. According to their segmentation method, we can approximately categorize them into region-based segmentation, data clustering, and edge-base segmentation. Clustering data is a fundamental task in machine learning. Given a set of data instances, the goal is to group them in a meaningful way, with the interpretation of the grouping dictated by the domain. In this paper we present a method for noise removal that makes use of graph clustering using both features (pixel intensity and connectivity). In this paper hierarchical clustering as well as centroid-based clustering is used. This will give us clusters and by analyzing these clusters we can identify noisy clusters. This is better method over the standard Vector Median Filter (VMF) [4]when noise ratio is high.

1. Jan Carlo Barca and Grace Rumantir, A Modified K-means Algorithm for Noise Reduction in Optical Motion Capture Data IEEE, ISBN: 0-7695-2841-4
2. Yu-Hsiang Wang (???),Segmentation Tutorial Graduate Institute of Communication Engineering National Taiwan University, Taipei, Taiwan, ROC
3. Satu Elisa Schaeffer, Graph clustering Computer Science review1 ( 2007) 27-64
4. J. Astola, P. Haavisto, and Y. Neuvo. Vector median filters, Proceedings of the IEEE, 74(4):678–689, April 1990
5. Olivier Lezoray, Vinh Thong Ta, Abderrahim Elmoataz, Impulse Noise Removal by Spectral Clustering and Regularization on Graphs, ANR-06-MDCA-008-01/FOGRIMM
6. Ying Huang Mei Xie , Weisheng Li , Lifang Zhou, Research on K-means Median Filter,Journal of Information & Computational Science 8: 6 (2011) 961–968
7. R. Lukac and K. Plataniotis. A taxonomy of color image filtering and enhancement solutions ,Advances in Imaging and Electron Physics, volume 140,pages 187–264. Elsevier, 2006.
8. R. H. Chan, C. -W. Ho, and M. Nikolova. Salt-and-pepper noise removal by median-type noise detectors and detailpreserving regularization. IEEE Transactions on Image Processing, 14(10):1479–1485, 2005.
9. Ali Ridho Barakbah and Yasushi Kiyoki, A New Approach for Image Segmentation using Pillar-Kmeans Algorithm, International Journal of Information and Communication Engineering 6:2 2010
10. B. F. Momin, P. M. Yelmar, Modifications in K-Means Clustering Algorithm, International Journal of Soft Computing and Engineering (IJSCE) ISSN: 2231-2307, Volume-2, Issue-3, July 2012

Graph Clustering Image Processing Vmf Noise Removal Hierarchical Clustering Centroid-based Clustering K-means Clustering