25-10-2012, 05:41 PM
Determining the number of cluster and Color Image Segmentation using K means
ABSTRACT
Clustering is the unsupervised classification of patterns (observations, data items, or feature vectors) into groups (clusters). The clustering problem has been addressed in many contexts and by researchers in many disciplines; this reflects its broad appeal and usefulness as one of the steps in exploratory data analysis. However, clustering is a difficult problem combinatorially, and differences in assumptions and contexts in different communities have made the transfer of useful generic concepts and methodologies slow to occur. Image segmentation is also another attractive field for research as Extraction of Region of Interest (ROI) from geographical map image is a very important field recently. When image segmentation is treated as a clustering of color pixels, K-means clustering became famous to cluster the color space samples. This paper presents an overview of existing clustering algorithms, with a goal of providing useful advice and references to fundamental concepts accessible to the broad community of clustering practitioners. Determining the number of cluster is a running burning question for clustering. We tried to propose a method for clustering validation. We present taxonomy of clustering techniques, and identify crosscutting themes and recent advances. We also describe a big famous filed of application of clustering- Image segmentation