Karmakar, G. C. and Dooley, L. S.
|DOI (Digital Object Identifier) Link:||http://doi.org/10.1016/S0167-8655(02)00069-7|
|Google Scholar:||Look up in Google Scholar|
Fuzzy rule based image segmentation techniques tend in general, to be application dependent with the structure of the membership functions being predefined and in certain cases, the corresponding parameters being manually determined. The net result is that the overall performance of the segmentation technique is very sensitive to parameter value selections. This paper addresses these issues by introducing a generic fuzzy rule based image segmentation (GFRIS) algorithm, which is both application independent and exploits inter-pixel spatial relationships. The GFRIS algorithm automatically approximates both the key weighting factor and threshold value in the definitions of the fuzzy rule and neighbourhood system, respectively. A quantitative evaluation is presented between the segmentation results obtained using GFRIS and the popular fuzzy c-means (FCM) and possibilistic c-means (PCM) algorithms. The results demonstrate that GFRIS exhibits a considerable improvement in performance compared to both FCM and PCM, for many different image types.
|Item Type:||Journal Article|
|Academic Unit/Department:||Faculty of Science, Technology, Engineering and Mathematics (STEM) > Computing and Communications
Faculty of Science, Technology, Engineering and Mathematics (STEM)
|Interdisciplinary Research Centre:||Centre for Research in Computing (CRC)|
|Depositing User:||Laurence Dooley|
|Date Deposited:||19 Aug 2008 06:37|
|Last Modified:||04 Oct 2016 10:11|
|Share this page:|