Type-2 kernelized fuzzy c-means algorithm based on the uncertain width of Gaussian kernel with applications in MR image segmentation

DOI

10.2991/ic3me-15.2015.259How to use a DOI?

Keywords

KFCM; Type-2 fuzzy sets; Gaussian kernel; MR image.

Abstract

While fuzzy c-means is a popular soft clustering method, its effectiveness is largely limited to spherical clusters. By applying kernel tricks, the kernel fuzzy c-means algorithm attempts to address this problem by mapping data with nonlinear relationships to appropriate feature spaces. Kernel width is crucial for effective kernel clustering. Unfortunately, for most applications, it is not easy to find the right width. To design and manage the uncertainty for kernel width, we propose a type-2 kernelized fuzzy c-means algorithm (T2KFCM). We extend the type-1 fuzzy sets of membership to interval type-2 fuzzy sets using two widths and which creates a footprint of uncertainty for the membership. Experiments on MR (Magnetic Resonance) image are given that compare kernelized FCM (KFCM) with T2KFCM. The results show that T2KFCM compares favorably to both of the previous models.

Copyright

© 2015, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Qinli Zhang
AU  - Yajun Bi
AU  - Zhigang Gong
PY  - 2015/08
DA  - 2015/08
TI  - Type-2 kernelized fuzzy c-means algorithm based on the uncertain width of Gaussian kernel with applications in MR image segmentation
BT  - Proceedings of the 3rd International Conference on Material, Mechanical and Manufacturing Engineering
PB  - Atlantis Press
SP  - 1351
EP  - 1354
SN  - 2352-5401
UR  - https://doi.org/10.2991/ic3me-15.2015.259
DO  - 10.2991/ic3me-15.2015.259
ID  - Zhang2015/08
ER  -