文章基本信息

标题：Image Segmentation Based on a Finite Generalized New Symmetric Mixture Model with K Means
本地全文：下载
作者：M Seshashayee ; K Srinivas Rao ; Ch Satyanarayana 等
期刊名称：International Journal of Computer Science Issues
印刷版ISSN：1694-0784
电子版ISSN：1694-0814
出版年度：2011
卷号：8
期号：3
出版社：IJCSI Press
摘要：In this paper a novel image segmentation and retrieval method based on finite new symmetric mixture model with K-means clustering is developed. Here it is considered that pixel intensities in each image region follow a new symmetric distribution. The new symmetric distribution includes platy-kurtic and meso-kurtic distributions. This also includes Gaussian mixture model as a particular case. The number of components (image regions) is obtained through K-means algorithm. The model parameters are estimated by deriving the updated equations of the EM algorithm. The segmentation of the image is done by maximizing the component likelihood. The performance of the proposed algorithm is studied by computing the segmentation performance metrics like, PRI, VOI, and GCE. The ability of this method for image retrieval is demonstrated by computing the image quality metrics for five images namely HORSE, MAN, BIRD, BOAT and TOWER. The experimental results show that this method outperforms the existing model based image segmentation methods.
关键词：Image segmentation; EM algorithm; New Symmetric Distribution. Image Quality Metrics