Alexandre L. M. Levada
ABSTRACT
This paper addresses the problem of maximum pseudo-likelihood estimation of the non-homogeneous Potts image model parameters using higher-order non-causal neighborhood systems in a computationally efficient way. The motivation is the development of a new methodology for contextual classification that uses combination of sub-optimal MRF algorithms for multispectral image classification, which requires accurate parameters estimation. Our objective is to make multispectral image contextual classification fully operational without human intervention. The results show that the method is consistent with real data and in the presence of random noise.
- Brent R. Algorithms for minimization without derivatives, Prentice Hall, New York, 1973.Google Scholar
- Cruvinel, P. E., Cesareo, R., Mascarenhas, S. "X and γ-rays computerized minitomograph scanner for soil science", IEEE Trans. on Instrumentation and Measurements, v. 39, n. 5, 745--750, 1990.Google ScholarCross Ref
- Dubes, R., Jain, A. "Random field models in image analysis", Jounrnal of Applied Statistics, v. 16, n. 2, 131--164, 1989.Google ScholarCross Ref
- Ising, E. "Beitrag zur Theorie des Ferromagnetismus", Z. Physik, 31, pp. 253--258, 1925 (German).Google ScholarCross Ref
- Li, S. Z. Markov Random Field modeling in image analysis, Springer, Second Edition, Tokyo, 2001. Google ScholarDigital Library
- Mascarenhas, N. D. A., Frery, A. C. "SAR image filtering with the ICM algorithm", International Geoscience and Remote Sensing Symposium, v. 4, 2185--2187, 1994.Google ScholarCross Ref
- Solberg, A. H. S. "Flexible nonlinear contextual classification", Patten Recognition Letters, v. 25, 1501--1508, 2004. Google ScholarDigital Library
- Waks, A., Tretiak, O. J., Gregoriou, G. K. "Restoration of noisy regions modeled by noncausal Markov random fields of unknown parameters", International Conference on Pattem Recognition (ICPR), v. 2, 170--175, 1990.Google ScholarCross Ref
- Won, C. S. Gray, R. M. Stochastic Image Processing, Kluwer Academics/Plenum Publishers, New Yourk, 2004.Google ScholarCross Ref
- Wu, J., Chung, A. C. S. "A segmentation model using compound Markov Random Fields based on a boundary model", IEEE Transactions on Image Processing, v. 16, n. 1, 241--252, 2007. Google ScholarDigital Library
- Yamazaki, T., Gingras, D. "Image classification using spectral and spatial information based on MRF Models", IEEE Transactions on Image Processing, v. 4, n. 9, 1333--1339, 1995. Google ScholarDigital Library
Recommendations
MAP-MRF super-resolution image reconstruction using maximum pseudo-likelihood parameter estimation
ICIP'09: Proceedings of the 16th IEEE international conference on Image processingIn this paper, we address the parameter estimation of a Super-Resolution Image Reconstruction approach following a Maximum a Posteriori Probability (MAP) algorithm. The Generalized Isotropic Multi-Level Logistic (GIMLL) Markov Random Field (MRF) model ...
A novel MAP-MRF approach for multispectral image contextual classification using combination of suboptimal iterative algorithms
In this paper we present a novel approach for multispectral image contextual classification by combining iterative combinatorial optimization algorithms. The pixel-wise decision rule is defined using a Bayesian approach to combine two MRF models: a ...
Improving Potts MRF Model Parameter Estimation in Image Analysis
CSE '08: Proceedings of the 2008 11th IEEE International Conference on Computational Science and EngineeringThis paper presents a novel pseudo-likelihood equation for the estimation of the Potts MRF model parameter on second-order neighborhood systems. Experiments with simulated images comparing the proposed estimation method with a recent maximum likelihood ...
Comments