New Feature Extraction Technique for Color Image Clustering
Manish Maheshwari, Mahesh Motwani, Sanjay Silakari · 2013
The fundamental data clustering problem may be defined as the process of grouping the data objects into classes or clusters, so that objects within a cluster have high similarity in comparison to one another but are very dissimilar to objects in other clusters. This paper produces an efficient new model for grouping of color images. A new color quantization ordering scheme that focuses on color as feature and considers Hue-Value and Saturation (HVS) space is proposed. Image pixel color is quantized into 54 colors and histogram of these 54 colors is calculated. To form clusters of images k-means algorithm is applied. Clustering analyzes data objects without consulting a known class label. In general, the class labels are not present in the training data simply because they are not known to begin with. Clustering can be used to generate such labels. The objects are clustered or grouped based on the principle of maximizing the intraclass similarity and minimizing the interclass similarity. (4) (5). In this paper we propose a data mining approach to cluster the images based on color feature. The concept of color histogram is used to obtain the features. RGB color space is converted to Hue, Saturation and Value (HSV) color space. Based on Hue, Saturation and Value, image is quantized to 54 colors and histogram of these 54 colors is formed. A K-means clustering algorithm is applied to cluster the images. The rest of the paper is organized as follows: In section two we provide an overview of image retrieval, histogram, HSV color and clustering. In section three we present the rules for quantization of the HSV color model and the calculation of histogram values. Experiments and results of clustering algorithms are discussed in section four.