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：2
页码：1248-1252
出版社：TechScience Publications
摘要：Segmentation of the images is often required as a preliminary and indispensable stage in the computer aided medical image process particularly during the clinical analysis of magnetic resonance (MR) brain image. K-means, Fuzzy c-means (FCM) clustering algorithm has been used in medical image segmentations, but the disadvantage of the k-means algorithm is weak pixel assignment could occur if the pixel with the equal minimum Euclidean distance to two or more adjacent cluster and it may be assigned to the higher variance cluster leading to dead center problems .To overcome that problem, the soft membership based called the Fuzzy cmeans( FCM) clustering algorithm is proposed .Fuzzy clustering using fuzzy c-means (FCM) algorithm proved to be superior over the other clustering approaches in terms of segmentation efficiency. But the disadvantage of the FCM algorithm is the large computational required for convergence and it is sensitive to noise because if not taking into account the spatial information. To overcome the above problem a modified FCM algorithm, for MRI brain image segmentation is presented in this paper. A comparative feature vector space is used for the segmentation technique. Comparative analysis in terms if segmentation efficiency and convergence rate is performed between the conventational FCM and the modified FCM. Experimental results show prior results for the modified FCM algorithm in terms of the performance measure.
