Please use this identifier to cite or link to this item:
Full metadata record
DC FieldValueLanguage
dc.contributor.authorNanda, P K-
dc.contributor.authorPatra, D-
dc.contributor.authorPradhan, A-
dc.identifier.citationProceedings of the 2nd Indian International Conference on Artificial Intelligence (IICAI 2005),en
dc.descriptionCopyright for this article belongs to proceedings publisheren
dc.description.abstractIn this paper, we propose a hybrid Tabu Expectation Maximization (TEM) Algorithm for segmentation of Brain Magnetic Resonance (MR) images in both supervised and unsupervised framewrok. Gaussian Hidden Markov Random Field (GHMRF) is used to model the available degraded image. In supervised framework, the apriori image MRF model parameters as well as the GHMRF model parameters are assumed to be known. The class labels are estimated using the Maximum a Posteriori (MAP) estimation criterion. In unsupervised framework, the problem of model parameter estimation and label estimation is formulated in Expectation Maximization (EM) framework. The labels are estimated using the proposed Tabu Search algorithm while the model parameters are the maximum likelihood estimates. Our proposed algorithm yields results with arbitrary initial paramater set and thus overcomes the problem of proper choice of initial parameters. The results obtained are comparable with the results obtained by using the algorithm proposed by Zhang [15] , where the Iterated Conditional Mode (ICM) algorithm is used for computing the MAP estimates.en
dc.format.extent857360 bytes-
dc.titleBrain MR Image Segmentation Using Tabu Search and Hidden Markov Random Field Modelen
Appears in Collections:Conference Papers

Files in This Item:
File Description SizeFormat 
paper_aiconf_pune.pdf837.27 kBAdobe PDFView/Open

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.