Fuzzy Expectation Maximum Algorithm for Magnetic Resonance Image Segmentation

Abstract:

Article Preview

Fuzzy clustering algorithm especially the fuzzy c-means (FCM) algorithm has been widely used for segmentation of brain magnetic resonance (MR) images. However, the conventional FCM algorithm has a very serious shortcoming, i.e., the algorithm tends to balance the number of points in each cluster during the classification. Therefore, when this algorithm is applied to segment the MR images with quite different size of objects, it will lead to bad segmentation. To overcome this problem, a novel fuzzy expectation maximization (FEM) algorithm is presented in this paper. The algorithm is developed by extending the classical hard EM algorithm into soft EM algorithm through integrating the fuzzy and statistical techniques. Compared with the FCM algorithm, the experimental results on MR image segmentation clearly indicate that the proposed FEM algorithm has better performance for the segmentation.

Info:

Periodical:

Key Engineering Materials (Volumes 439-440)

Edited by:

Yanwen Wu

Pages:

1618-1623

DOI:

10.4028/www.scientific.net/KEM.439-440.1618

Citation:

Y. Yang "Fuzzy Expectation Maximum Algorithm for Magnetic Resonance Image Segmentation", Key Engineering Materials, Vols. 439-440, pp. 1618-1623, 2010

Online since:

June 2010

Authors:

Export:

Price:

$35.00

In order to see related information, you need to Login.

In order to see related information, you need to Login.