
New Method to Optimize Initial Point Values of Spatial Fuzzy c-means Algorithm
Author(s) -
Iman Omidvar Tehrani,
Subariah Ibrahim,
Habib Haron
Publication year - 2015
Publication title -
international journal of electrical and computer engineering
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.277
H-Index - 22
ISSN - 2088-8708
DOI - 10.11591/ijece.v5i5.pp1035-1044
Subject(s) - initialization , computer science , algorithm , image segmentation , cluster analysis , histogram , segmentation , convergence (economics) , fuzzy logic , artificial intelligence , image (mathematics) , economics , programming language , economic growth
Fuzzy based segmentation algorithms are known to be performing well on medical images. Spatial fuzzy C-means (SFCM) is broadly used for medical image segmentation but it suffers from optimum selection of seed point initialization which is done either manually or randomly. In this paper, an enhanced SFCM algorithm is proposed by optimizing the SFCM initial point values. In this method in order to increasing the algorithm speed first the approximate initial values are determined by calculating the histogram of the original image. Then by utilizing the GWO algorithm the optimum initial values could be achieved. Finally By using the achieved initial values, the proposed method shows the significant improvement in segmentation results. Also the proposed method performs faster than previous algorithm i.e. SFCM and has better convergence. Moreover, it has noticeably improved the clustering effect.