A markov random field based hybrid algorithm with simulated annealing and genetic algorithm for image segmentation

Authors:
Xinyu Du;Yongjie Li;Wufan Chen;Yi Zhang;Dezhong Yao
Affiliations:
Center of NeuroInformatics, School of Life Science & Technology, University of Electronic Science and Technology of China, Chengdu, PR China;Center of NeuroInformatics, School of Life Science & Technology, University of Electronic Science and Technology of China, Chengdu, PR China;Center of NeuroInformatics, School of Life Science & Technology, University of Electronic Science and Technology of China, Chengdu, PR China;Center of NeuroInformatics, School of Life Science & Technology, University of Electronic Science and Technology of China, Chengdu, PR China;Center of NeuroInformatics, School of Life Science & Technology, University of Electronic Science and Technology of China, Chengdu, PR China
Venue:
ICNC'06 Proceedings of the Second international conference on Advances in Natural Computation - Volume Part I
Year:
2006

Citing 5
Cited 1

Texture segmentation based on a hierarchical Markov random field model

Signal Processing
Markov random field modeling in image analysis

Markov random field modeling in image analysis
Classification and Clustering for Knowledge Discovery (Studies in Computational Intelligence)

Classification and Clustering for Knowledge Discovery (Studies in Computational Intelligence)
Image segmentation using evolutionary computation

IEEE Transactions on Evolutionary Computation
Evolutionary optimization with Markov random field prior

IEEE Transactions on Evolutionary Computation

IR image segmentation using GA-MRF with neighborhood labels coding

CAR'10 Proceedings of the 2nd international Asia conference on Informatics in control, automation and robotics - Volume 2

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper, a simulated algorithm-genetic (SA-GA) hybrid algorithm based on a Markov Random Field (MRF) model (MRF-SA-GA) is introduced for image de-noising and segmentation. In this algorithm, a population of potential solutions is maintained at every generation, and for each solution a fitness value is calculated with a fitness function, which is constructed based on the MRF potential function according to Metropolis algorithm and Bayesian rule. Two experiments are selected to verify the performance of the hybrid algorithm, and the preliminary results show that MRF-SA-GA outperforms SA and GA alone.