gunawan, Wawan and arifin, agus zainal and Rosidin, Undang and Kadaritna, Nina (2019) Spatial Condition in Intuitionistic Fuzzy C-Means Clustering for Segmentation of Teeth in Dental Panoramic Radiographs. Indonesian Journal of Computing and Cybernetics Systems), 13 (4). ISSN 2460-7258

[img] Text
51. Artikel Wawan dkk.pdf

Download (358kB)
Official URL: http://www.cscjournals.org/journals/IJCSS/descript...

Abstract

Dental panoramic radiographs heavily depend on the performance of the segmentation method due to the presence of unevenly illumination and low contrast of the images. Conditional Spatial Fuzzy C-mean (csFCM) Clustering have been proposed to achieve through the incorporation of the component and added in the FCM to cluster grouping. This algorithm directs with consideration conditioning variables that consider membership value. However, csFCM does not consider Intuitionistic Fuzzy Set to take final membership and final nonmembership value into account, the effect does not wipe off the deviation by illumination and low contrast of the images completely for improvement to skip some scope. In this current paper, we introduced a new image segmentation method namely Conditional Spatial in Intuitionistic Fuzzy C-Means Clustering for Segmentation of Teeth in Dental Panoramic Radiographs. Our proposed method adds hesitation function aiming to settle the indication of the knowledge lack that belongs to the final membership function to get a better segmentation result. The experiment result shows this method achieves better segmentation performance with misclassification error (ME) and relative foreground area error (RAE) values are 4.77 and 4.27 respectively.

Item Type: Article
Subjects: Q Science > Q Science (General)
T Technology > T Technology (General)
Divisions: Fakultas Keguruan dan Ilmu Pendidikan (FKIP) > Prodi Magister Pendidikan Fisika
Depositing User: Dr. Undang Rosidin
Date Deposited: 18 Nov 2019 07:08
Last Modified: 18 Nov 2019 07:08
URI: http://repository.lppm.unila.ac.id/id/eprint/16716

Actions (login required)

View Item View Item