Title :
A robust segmentation method for the AFCM-MRF model in noisy image
Author :
Tam, Simon C F ; Leung, C.C. ; Tsui, W.K.
Abstract :
A robust image segmentation algorithm based on Alternative Fuzzy C-mean clustering algorithm (AFCM) with Markov Random Field (MRF) is presented in this paper. Due to disregard of spatial constraint information, the results using Fuzzy C-Mean (FCM) and AFCM are corrupted by noise. In order to improve the robustness of noise, the spatial constraint information of an image is represented by MRF with the Gibbs function which is based on the AFCM. Comparison to the FCM, AFCM, FCM-MRF model, and the proposed algorithm has been demonstrated by the simulation and real images. Results show that AFCM-MRF model achieves better performance than other methods.
Keywords :
Markov processes; fuzzy set theory; image representation; image segmentation; pattern clustering; random processes; AFCM-MRF model; Gibbs function; Markov random field; alternative fuzzy C-mean clustering algorithm; image representation; robust noisy image segmentation method; spatial constraint information; Bayesian methods; Clustering algorithms; Computer vision; Constraint optimization; Equations; Image segmentation; Machine learning algorithms; Markov random fields; Noise robustness; Pixel;
Conference_Titel :
Fuzzy Systems, 2009. FUZZ-IEEE 2009. IEEE International Conference on
Conference_Location :
Jeju Island
Print_ISBN :
978-1-4244-3596-8
Electronic_ISBN :
1098-7584
DOI :
10.1109/FUZZY.2009.5277193
