期刊名称：International Journal of Computer Science and Information Technologies
电子版ISSN：0975-9646
出版年度：2015
卷号：6
期号：1
页码：514-518
出版社：TechScience Publications
摘要：In this research work we have studied Clustering k-means algorithm and Lattice Boltzmann Method. Image segmentation has proved its applicability in various areas like satellite figure processing, medical image processing and many more. In the present development the researchers tries to develop hybrid image segmentation techniques to generates efficient segmentation. Due to the progress of the parallel programming, the lattice Boltzmann method (LBM) has attracted much attention as a fast alternative approach for solving partial differential equations. This idea leads to provide color image segmentation using single channel segments of multichannel images. Though this method is widely adopted but doesn’t provide complete true segmentation of multichannel i.e. color images because a color image contains three different channels for Red, green and blue components. Hence segmenting a color image, by having only single channel segments information, will definitely loose important segment regions of color images. To overcome this problem this paper work starts with the development of Enhanced Level Set Segmentation for single channel Images Using k-means Clustering algorithm and Lattice Boltzmann Method. Data clustering is one of the important data mining methods. It is a procedure of discovery classes of a data set with most similarity in the same class and most dissimilarity between different classes .In this paper we study on hard clustering K-Means algorithm that is mostly based on Euclidean distance measure [ 26].Hard clustering methods are based on traditional set theory, and require that an object either does or does not belong to a bunch. Hard clustering means partitioning the data into a specified number of mutually exclusive subsets. [27]. For the study of the proposed segmentation scheme three segmentation parameters have been utilized, they are Probabilistic Rand Index (PRI), Variation of Information (VOI) and Global Consistency Error (GCE).
