DocumentCode :
2649855
Title :
A Selective Fuzzy Region Competition Model for Multiphase Image Segmentation
Author :
Borges, Vinicius R Pereira ; Barcelos, Celia A Zorzo ; Guliato, Denise ; Batista, Marcos Aurélio
Author_Institution :
Comput. & Math. Fac., INCT-MACC, Fed. Univ. of Uberlandia, Uberlandia, Brazil
fYear :
2011
fDate :
7-9 Nov. 2011
Firstpage :
118
Lastpage :
125
Abstract :
This paper presents a multiphase image segmentation model based on Fuzzy Region Competition. The proposed model approximates image regions by probability density functions and uses a supervised approach in the segmentation process. The strategy of the proposed model is to perform two-phase Fuzzy Region Competition model several times, where a hard partition is obtained in each round from the determined fuzzy membership function. Consequently, the segmentation process is soft, while the final result is hard, given the simplicity of avoiding non-overlapping and vacuum regions. The proposed model was validated using multiphase images, which showed to be robust under the presence of noise and presented good accuracy when dealing with texturized and natural images.
Keywords :
approximation theory; fuzzy set theory; image segmentation; image texture; natural scenes; probability; fuzzy membership function; image region approximation; multiphase image segmentation; natural images; nonoverlapping regions; probability density functions; selective fuzzy region competition model; supervised approach; texturized images; two-phase fuzzy region competition model; vacuum regions; Computational modeling; Image segmentation; Level set; Mathematical model; Minimization; Probability density function; Projection algorithms; Fuzzy Region Competition; Multiphase Segmentation; Probability Density Functions; Variational Methods;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Tools with Artificial Intelligence (ICTAI), 2011 23rd IEEE International Conference on
Conference_Location :
Boca Raton, FL
ISSN :
1082-3409
Print_ISBN :
978-1-4577-2068-0
Electronic_ISBN :
1082-3409
Type :
conf
DOI :
10.1109/ICTAI.2011.26
Filename :
6103315
Link To Document :
بازگشت