SAR Image Segmentation Based on Bayesian Network

Abstract:

In this paper, we propose a Bayesian network model. Firstly, the Bayesian network model is introduced, and Belief Propagation (BP) algorithm is utilized for model estimation. Then ExpectationMaximization (EM) algorithm is used for parameter estimation of the Bayesian network. Finally, the SAR image is segmented by calculating the Maximum Posteriori Probability (MAP) of each pixel. Experimental results show that, comparing with the Markov Random Field - Intersecting Cortical Model (MRF-ICM), our Bayesian network model gives better results in both segmentation and time-consuming.

• Periodical: Advanced Materials Research (Volumes 756-759)
• Pages: 1835-1839
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.756-759.1835
• Online since: September 2013
• Authors: Jian Guang Zhang, Yong Xia Li, Ping Chen
• Keywords: Belief Propagation, Expectation Maximization, Maximum Posteriori Probability

References

[1] Michael I. Jordan. Graphical Models, Computer Science Division and Department of Statistics University of California.

[2] Yang Yong, Cao Yongfeng, Sun Hong. Segmentation of SAR imagery using the Gaussian Markov random field model, IEEE Trans. International Conference on Signal Processing, (ICSP), Volume 3, Page(s): 1977 - 1980, (2004).

DOI: 10.1109/icosp.2004.1442160

[3] Xiaodong Lu, Fengqi Zhou, Jun Zhou. Synthetic aperture radar image segmentation based on improved fuzzy Markov random field model, IEEE Trans. International Symposium on Systems and Control in Aerospace and Astronautics (ISSCAA), Page(s): 4 pp. - 1208, (2006).

DOI: 10.1109/isscaa.2006.1627582

[4] Lankoande, O., Hayat, M.M., Balu Santhanam. Segmentation of SAR images based on Markov random field model, IEEE International Conference on Systems, Man and Cybernetics, Volume 3, Page(s): 2956 - 2961, (2005).

DOI: 10.1109/icsmc.2005.1571599

[5] Tupin, F. Markovian methods for SAR image processing, IEEE International Geoscience and Remote Sensing Symposium(IGARSS ), Volume 3, Page(s): 1164 – 1166, (2001).

DOI: 10.1109/igarss.2001.976779

[6] Fei Liu, Dongxiang Xu, Chun Yuan, Kerwin, W. Image segmentation based on Bayesian network-Markov random field model and its application to in vivo plaque composition, IEEE International Symposium, April 2006 Page(s): 141 – 144.

DOI: 10.1109/isbi.2006.1624872

[7] Mortensen, E., Jin Jia. A Bayesian Network Framework for Real-Time Object Selection, IEEE Conference on Computer Vision and Pattern Recognition Workshop, Page(s): 44 - 44, June, (2004).

DOI: 10.1109/cvpr.2004.278

[8] Shih-Shinh Huang, Li-Chen Fu, Pei-Yung Hsiao. A Bayesian Framework for Foreground Segmentation, IEEE International Conference on Systems, Man and Cybernetics, Volume 3, Page(s): 1980 - 1985, (2006).

DOI: 10.1109/icsmc.2006.385021

[9] Zhang Cui, Li Su-dan, Wang Zheng-zhi. SAR image segmentation based on Markov random field model. Remote Sensing Technology and Application, 2001, 16(1): 66–68.

[10] Bill Freeman, Fredo Durand, Graphical models, belief propagation, and Markov random fields, ppt(MIT), March 21, (2005).

[11] Michael K. NG. K-means-type algorithms on distributed memory computer [J]. International Journal of High speed Computing. 2000, 11(2): 75-91.

DOI: 10.1142/s0129053300000096