Unsupervised Segmentation of Nonstationary Data using Triplet Markov Chains

Mohamed El Yazid Boudaren, Emmanuel Monfrini, Kadda Beghdad Bey, Ahmed Habbouchi, Wojciech Pieczynski

Abstract

An important issue in statistical image and signal segmentation consists in estimating the hidden variables of interest. For this purpose, various Bayesian estimation algorithms have been developed, particularly in the framework of hidden Markov chains, thanks to their efficient theory that allows one to recover the hidden variables from the observed ones even for large data. However, such models fail to handle nonstationary data in the unsupervised context. In this paper, we show how the recent triplet Markov chains, which are strictly more general models with comparable computational complexity, can be used to overcome this limit through two different ways: (i) in a Bayesian context by considering the switches of the hidden variables regime depending on an additional Markov process; and, (ii) by introducing Dempster-Shafer theory to model the lack of precision of the hidden process prior distributions, which is the origin of data nonstationarity. Furthermore, this study analyzes both approaches in order to determine which one is better-suited for nonstationary data. Experimental results are shown for sampled data and noised images.

References

  1. Ait-El-Fquih, B. and Desbouvries, F. (2005). Bayesian smoothing algorithms in pairwise and triplet markov chains. In Statistical Signal Processing, 2005 IEEE/SP 13th Workshop on, pages 721-726. IEEE.
  2. Ait-El-Fquih, B. and Desbouvries, F. (2006). Kalman filtering in triplet Markov chains. Signal Processing, IEEE Transactions on, 54(8):2957-2963.
  3. An, L., Li, M., Boudaren, M. E. Y., and Pieczynski, W. (2016). Evidential correlated gaussian mixture markov model for pixel labeling problem. In International Conference on Belief Functions, pages 203- 211. Springer.
  4. Bardel, N. and Desbouvries, F. (2012). Exact Bayesian prediction in a class of Markov-switching models. Methodology and Computing in Applied probability, 14(1):125-134.
  5. Baum, L. E., Petrie, T., Soules, G., and Weiss, N. (1970). A maximization technique occurring in the statistical analysis of probabilistic functions of Markov chains. The annals of mathematical statistics, pages 164-171.
  6. Benboudjema, D. and Pieczynski, W. (2007). Unsupervised statistical segmentation of nonstationary images using triplet Markov fields. Pattern Analysis and Machine Intelligence, IEEE Transactions on, 29(8):1367-1378.
  7. Blanchet, J. and Forbes, F. (2008). Triplet Markov fields for the classification of complex structure data. Pattern Analysis and Machine Intelligence, IEEE Transactions on, 30(6):1055-1067.
  8. Boudaren, M. E. Y., An, L., and Pieczynski, W. (2016a). Dempster-Shafer fusion of evidential pairwise Markov fields. Approximate Reasoning, International Journal of, 74:13-29.
  9. Boudaren, M. E. Y., An, L., and Pieczynski, W. (2016b). Unsupervised segmentation of SAR images using Gaussian mixture-hidden evidential Markov fields. IEEE Geoscience and Remote Sensing Letters, 13(12):1865-1869.
  10. Boudaren, M. E. Y., Monfrini, E., and Pieczynski, W. (2011). Unsupervised segmentation of switching pairwise Markov chains. In Image and Signal Processing and Analysis (ISPA), 2011 7th International Symposium on, pages 183-188. IEEE.
  11. Boudaren, M. E. Y., Monfrini, E., and Pieczynski, W. (2012a). Unsupervised segmentation of nonstationary pairwise Markov chains using evidential priors. In EUSIPCO, pages 2243-2247, Bucharest. IEEE.
  12. Boudaren, M. E. Y., Monfrini, E., and Pieczynski, W. (2012b). Unsupervised segmentation of random discrete data hidden with switching noise distributions. IEEE Signal Process. Lett., 19(10):619-622.
  13. Boudaren, M. E. Y., Monfrini, E., Pieczynski, W., and Aïssani, A. (2012c). Dempster-Shafer fusion of multisensor signals in nonstationary Markovian context. EURASIP J. Adv. Sig. Proc., 2012:134.
  14. Boudaren, M. E. Y., Monfrini, E., Pieczynski, W., and Aissani, A. (2014). Phasic triplet Markov chains. IEEE Transactions on Pattern Analysis and Machine Intelligence, 36(11):2310-2316.
  15. Boudaren, M. E. Y. and Pieczynski, W. (2016a). DempsterShafer fusion of evidential pairwise Markov chains. Fuzzy Systems, IEEE Transactions on, 24(6):1598- 1610.
  16. Boudaren, M. E. Y. and Pieczynski, W. (2016b). Unified representation of sets of heterogeneous Markov transition matrices. Fuzzy Systems, IEEE Transactions on, 24(2):497-503.
  17. Bricq, S., Collet, C., and Armspach, J.-P. (2006). Triplet Markov chain for 3D MRI brain segmentation using a probabilistic atlas. In Biomedical Imaging: Nano to Macro, 2006. 3rd IEEE International Symposium on, pages 386-389. IEEE.
  18. Cappé, O., Moulines, E., and Rydén, T. (2005). Inference in hidden Markov models, volume 6. Springer, New York.
  19. Celeux, G., Chauveau, D., and Diebolt, J. (1996). Stochastic versions of the EM algorithm: an experimental study in the mixture case. Journal of Statistical Computation and Simulation, 55(4):287-314.
  20. Dehghani, H., Vahidi, B., Naghizadeh, R., and Hosseinian, S. (2013). Power quality disturbance classification using a statistical and wavelet-based hidden Markov model with Dempster-Shafer algorithm. International Journal of Electrical Power & Energy Systems, 47:368-377.
  21. Delmas, J. P. (1997). An equivalence of the EM and ICE algorithm for exponential family. IEEE transactions on signal processing, 45(10):2613-2615.
  22. Derrode, S. and Pieczynski, W. (2004). Signal and image segmentation using pairwise Markov chains. Signal Processing, IEEE Transactions on, 52(9):2477-2489.
  23. Forney Jr, G. D. (1973). The Viterbi algorithm. Proceedings of the IEEE, 61(3):268-278.
  24. Foucher, S., Germain, M., Boucher, J.-M., and Benie, G. B. (2002). Multisource classification using ICM and Dempster-Shafer theory. IEEE Transactions on Instrumentation and Measurement, 51(2):277-281.
  25. Fouque, L., Appriou, A., and Pieczynski, W. (2000). An evidential Markovian model for data fusion and unsupervised image classification. InInformation Fusion, 2000. FUSION 2000. Proceedings of the Third International Conference on, volume 1, pages TUB4-25. IEEE.
  26. Gan, L., Wu, Y., Liu, M., Zhang, P., Ji, H., and Wang, F. (2012). Triplet Markov fields with edge location for fast unsupervised multi-class segmentation of synthetic aperture radar images. IET image processing, 6(7):831-838.
  27. Habbouchi, A., Boudaren, M. E. Y., Aïssani, A., and Pieczynski, W. (2016). Unsupervised segmentation of markov random fields corrupted by nonstationary noise. IEEE Signal Processing Letters, 23(11):1607- 1611.
  28. Lanchantin, P., Lapuyade-Lahorgue, J., and Pieczynski, W. (2011). Unsupervised segmentation of randomly switching data hidden with non-Gaussian correlated noise. Signal Processing, 91(2):163-175.
  29. Lanchantin, P. and Pieczynski, W. (2005). Unsupervised restoration of hidden nonstationary Markov chains using evidential priors. Signal Processing, IEEE Transactions on, 53(8):3091-3098.
  30. Lapuyade-Lahorgue, J. and Pieczynski, W. (2006). Unsupervised segmentation of hidden semi-Markov non stationary chains. In Bayesian Inference and Maximum Entropy Methods in Science and Engineering(AIP Conference Proceedings Volume 872), volume 872, pages 347-354. Citeseer.
  31. Le Hégarat-Mascle, S., Bloch, I., and Vidal-Madjar, D. (1998). Introduction of neighborhood information in evidence theory and application to data fusion of radar and optical images with partial cloud cover. Pattern recognition, 31(11):1811-1823.
  32. Liu, G., Li, M., Wu, Y., Zhang, P., Jia, L., and Liu, H. (2014). Polsar image classification based on wishart TMF with specific auxiliary field.Geoscience and Remote Sensing Letters, IEEE, 11(7):1230-1234.
  33. McLachlan, G. and Krishnan, T. (2007). The EM algorithm and extensions, volume 382. John Wiley & Sons, New Jersey.
  34. Park, J., Chebbah, M., Jendoubi, S., and Martin, A. (2014). Second-order belief hidden markov models. In Belief Functions: Theory and Applications, pages 284-293. Springer.
  35. Pieczynski, W. (2003). Pairwise Markov chains. Pattern Analysis and Machine Intelligence, IEEE Transactions on, 25(5):634-639.
  36. Pieczynski, W. (2007). Multisensor triplet Markov chains and theory of evidence. International Journal of Approximate Reasoning, 45(1):1-16.
  37. Pieczynski, W. and Benboudjema, D. (2006). Multisensor triplet Markov fields and theory of evidence. Image and Vision Computing, 24(1):61-69.
  38. Pieczynski, W., Hulard, C., and Veit, T. (2003). Triplet Markov chains in hidden signal restoration. In International Symposium on Remote Sensing, pages 58-68. International Society for Optics and Photonics.
  39. Rabiner, L. (1989). A Tutorial on Hidden Markov Models and Selected Applications in Speech Recognition. Proceedings of the IEEE, 77(2):257-286.
  40. Ramasso, E. (2009). Contribution of belief functions to hidden Markov models with an application to fault diagnosis. In Proceedings of the IEEE International Workshop on Machine Learning for Signal Processing., pages 1-6.
  41. Reineking, T. (2011). Particle filtering in the DempsterShafer theory. International Journal of Approximate Reasoning, 52(8):1124-1135.
  42. Shafer, G. (1976). A mathematical theory of evidence, volume 1. Princeton university press Princeton, Princeton.
  43. Soubaras, H. (2010). On evidential Markov chains. In Foundations of reasoning under uncertainty, pages 247- 264. Springer, Berlin Heidelberg.
  44. Tupin, F., Bloch, I., and Maˆitre, H. (1999). A first step toward automatic interpretation of SAR images using evidential fusion of several structure detectors. Geoscience and Remote Sensing, IEEE Transactions on, 37(3):1327-1343.
  45. Wang, F., Wu, Y., Zhang, Q., Zhang, P., Li, M., and Lu, Y. (2013). Unsupervised change detection on SAR images using triplet Markov field model. Geoscience and Remote Sensing Letters, IEEE, 10(4):697-701.
  46. Wu, Y., Zhang, P., Li, M., Zhang, Q., Wang, F., and Jia, L. (2013). SAR image multiclass segmentation using a multiscale and multidirection triplet markov fields model in nonsubsampled contourlet transform domain. Information Fusion, 14(4):441-449.
  47. Yoji, Y., Tetsuya, M., Yoji, U., and Yukitaka, S. (2003). A method for preventing accidents due to human action slip utilizing HMM-based Dempster-Shafer theory. In Robotics and Automation, 2003. Proceedings. ICRA'03. IEEE International Conference on, volume 1, pages 1490-1496. IEEE.
  48. Zhang, P., Li, M., Wu, Y., Gan, L., Liu, M., Wang, F., and Liu, G. (2012a). Unsupervised multi-class segmentation of SAR images using fuzzy triplet Markov fields model. Pattern Recognition, 45(11):4018-4033.
  49. Zhang, P., Li, M., Wu, Y., Liu, M., Wang, F., and Gan, L. (2012b). SAR image multiclass segmentation using a multiscale TMF model in wavelet domain. Geoscience and Remote Sensing Letters, IEEE, 9(6):1099-1103.
Download


Paper Citation


in Harvard Style

Boudaren M., Monfrini E., Beghdad Bey K., Habbouchi A. and Pieczynski W. (2017). Unsupervised Segmentation of Nonstationary Data using Triplet Markov Chains . In Proceedings of the 19th International Conference on Enterprise Information Systems - Volume 1: ICEIS, ISBN 978-989-758-247-9, pages 405-414. DOI: 10.5220/0006276704050414


in Bibtex Style

@conference{iceis17,
author={Mohamed El Yazid Boudaren and Emmanuel Monfrini and Kadda Beghdad Bey and Ahmed Habbouchi and Wojciech Pieczynski},
title={Unsupervised Segmentation of Nonstationary Data using Triplet Markov Chains},
booktitle={Proceedings of the 19th International Conference on Enterprise Information Systems - Volume 1: ICEIS,},
year={2017},
pages={405-414},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0006276704050414},
isbn={978-989-758-247-9},
}


in EndNote Style

TY - CONF
JO - Proceedings of the 19th International Conference on Enterprise Information Systems - Volume 1: ICEIS,
TI - Unsupervised Segmentation of Nonstationary Data using Triplet Markov Chains
SN - 978-989-758-247-9
AU - Boudaren M.
AU - Monfrini E.
AU - Beghdad Bey K.
AU - Habbouchi A.
AU - Pieczynski W.
PY - 2017
SP - 405
EP - 414
DO - 10.5220/0006276704050414