Your browser does not support JavaScript!
http://iet.metastore.ingenta.com
1887

access icon free Hierarchical non-parametric Markov random field for image segmentation

  • Author(s): Xiangrong Wang 1, 2  and  Jieyu Zhao 1
    • View affiliations
  • Source: Volume 11, Issue 8, December 2017, p. 717 – 724
    DOI:  10.1049/iet-cvi.2016.0429 , Print ISSN 1751-9632, Online ISSN 1751-9640
© The Institution of Engineering and Technology
Received 07/12/2016, Accepted 07/08/2017, Revised 14/06/2017, Published 16/08/2017

Markov random fields (MRFs) are prominent in modelling image to handle image processing problems. However, they confront the bottleneck of model selection in further improving the performance. That is difficult to decide how many objects in an image automatically. Motivated by Bayesian non-parametric (BN) models, a layered BN MRF is proposed. The proposed model is hierarchical: the lower level is a random-field like model, while the higher level is a Chinese restaurant process (CRP). The clustering procedure can be formulated briefly as follows. The input data is first clustered with the lower level MRF to form a set of components. Then the higher level CRP is used to merge the components into larger clusters. Furthermore, a split–merge Monte Carlo Markov chain is employed. Quantitative evaluations over BSD500 data set and MSRC data set show the proposed model is comparable to the state-of-the-art BN models and other graphical models in modelling unsupervised distance-dependent problems.

Inspec keywords: nonparametric statistics; Markov processes; Bayes methods; image segmentation; random processes; Monte Carlo methods; pattern clustering

Other keywords: BSD500 data set; Bayesian nonparametric model; hierarchical nonparametric Markov random field; layered BN MRF; graphical models; clustering procedure; Chinese restaurant process; BN models; split-merge Monte Carlo Markov chain; image processing; random-field like model; model selection; quantitative evaluations; image segmentation; unsupervised distance-dependent problems; MSRC data set

Subjects: Markov processes; Optical, image and video signal processing; Monte Carlo methods; Markov processes; Computer vision and image processing techniques; Monte Carlo methods

References

    1. 1)
      • 14. Rahman, M.M., Horiguchi, S.: ‘A fully unsupervised color textured image segmentation algorithm using weighted mean histograms features’, Signal Image Video Process., 2012, 6, (2), pp. 197–209.
    2. 2)
      • 38. Martin, D., Fowlkes, C., Tal, D., et al: ‘A database of human segmented natural images and its application to evaluating segmentation algorithms and measuring ecological statistics’. Proc. Eighth IEEE Int. Conf. on Computer Vision, British Columbia, Canada, July 2001, pp. 416–423.
    3. 3)
      • 16. Ghosh, S., Ungureanu, A.B., Sudderth, E.B., et al: ‘Spatial distance dependent Chinese restaurant processes for image segmentation’, Adv. Neural Inf. Process. Syst., 2011, pp. 1476–1484.
    4. 4)
      • 7. Ahmadvand, A., Daliri, M.R., Zahiri, S.M.: ‘Segmentation of brain MR images using a proper combination of DCS based method with MRF’, Multimedia Tools Appl., 2017, https://doi.org/10.1007/s11042-017-4696-8.
    5. 5)
      • 2. Shi, J., Malik, J.: ‘Normalized cuts and image segmentation’, Comput. Vis. Pattern Recognit., 2000, 22, (8), pp. 888–905.
    6. 6)
      • 12. Schwarz, G.: ‘Estimating the dimension of a model’, Ann. Stat., 1978, 6, (2), pp. 461–464.
    7. 7)
      • 33. Martin, D.E., Fowlkes, C.C., Tal, D., et al: ‘A database of human segmented natural images and its application to evaluating segmentation algorithms and measuring ecological statistics’. IEEE 8th Int. Conf. on Computer Vision, 2001, vol. 2, pp. 416–423.
    8. 8)
      • 25. Li, C., Rana, S., Phung, D., et al: ‘Data clustering using side information dependent Chinese restaurant processes’, Knowl. Inf. Syst., 2016, 47, (2), pp. 463–488.
    9. 9)
      • 17. Polatkan, G., Zhou, M., Carin, L., et al: ‘A Bayesian nonparametric approach to image super-resolution’, IEEE Trans. Pattern Anal. Mach. Intell., 2015, 37, (2), pp. 346–358.
    10. 10)
      • 20. Nguyen, X., Gelfand, A.E.: ‘The Dirichlet labeling process for functional data analysis’. Technical Report, Citeseer, 2008.
    11. 11)
      • 5. Yin, S., Qian, Y., Gong, M.: ‘Unsupervised hierarchical image segmentation through fuzzy entropy maximization’, Pattern Recognit., 2017, 68, pp. 245–259.
    12. 12)
      • 13. Stanford, D.C., Raftery, A.E.: ‘Approximate Bayes factors for image segmentation: the pseudolikelihood information criterion (PLIC)’, IEEE Trans. Pattern Anal. Mach. Intell., 2002, 24, (11), pp. 1517–1520.
    13. 13)
      • 21. Chatzis, S.P., Tsechpenakis, G.: ‘The infinite hidden Markov random field model’, IEEE Trans. Neural Netw., 2010, 21, (6), pp. 1004–1014.
    14. 14)
      • 10. Ahmadvand, A., Daliri, M.R.: ‘Improving the runtime of MRF based method for MRI brain segmentation’, Appl. Math. Comput., 2015, 256, pp. 808–818.
    15. 15)
      • 31. Ishwaran, H., James, L.F.: ‘Gibbs sampling methods for stick-breaking priors’, J. Am. Stat. Assoc., 2001, 96, (453), pp. 161–173.
    16. 16)
      • 6. Li, A., Wang, X., Yan, K., et al: ‘Multilevel affinity graph for unsupervised image segmentation’. 2016 IEEE Int. Conf. on Image Processing (ICIP), 2016, pp. 1264–1268.
    17. 17)
      • 30. Xiao, H., Stibor, T.: ‘Efficient collapsed Gibbs sampling for latent Dirichlet allocation’, J. Mach. Learn. Res. Proc. Track, 2010, 13, pp. 63–78.
    18. 18)
      • 11. Barbu, A., Zhu, S.C.: ‘Generalizing Swendsen–Wang for image analysis’, J. Comput. Graph. Stat., 2007, 16, (4), pp. 877–900.
    19. 19)
      • 28. Ghosh, S., Raptis, M., Sigal, L., et al: ‘Nonparametric clustering with distance dependent hierarchies’. 30th Conf. on Uncertainty in Artificial Intelligence, Quebec, Canada, July, 2014.
    20. 20)
      • 27. Andalib, A., Babamir, S.M.: ‘A class-based link prediction using distance dependent Chinese restaurant process’, Physica A Stat. Mech. Appl., 2016, 456, pp. 204–214.
    21. 21)
      • 29. Wang, C., Blei, D.M.: ‘A split-merge MCMC algorithm for the hierarchical Dirichlet process’, arXiv preprint arXiv: 2012, 1201.1657.
    22. 22)
      • 36. Powers, D.M.: ‘Evaluation: from precision, recall and F-measure to ROC, informedness, markedness and correlation’, J. Mach. Learn. Technol., 2011, 2, (1), pp. 37–63.
    23. 23)
      • 4. Gao, L., Song, J., Nie, F., et al: ‘Graph-without-cut: an ideal graph learning for image segmentation’. Thirtieth AAAI Conf. on Artificial Intelligence, Phoenix, Arizona, USA, February 2016.
    24. 24)
      • 8. Ahmadvand, A., Sharififar, M., Daliri, M.R., et al: ‘Supervised segmentation of MRI brain images using combination of multiple classifiers’, Australas. Phys. Eng. Sci. Med., 2015, 38, (2), pp. 241–253.
    25. 25)
      • 15. Orbanz, P., Buhmann, J.M.: ‘Nonparametric Bayesian image segmentation’, Int. J. Comput. Vis., 2008, 77, (1–3), pp. 25–45.
    26. 26)
      • 37. Meilǎ, M.: ‘Comparing clusterings: an axiomatic view’. Proc. of the 22nd Int. Conf. on Machine Learning, 2005, pp. 577–584.
    27. 27)
      • 1. Boykov, Y., Lee, V.S., Rusinek, H., et al: ‘Segmentation of dynamic n-d data sets via graph cuts using Markov models’. Medical Image Computing and Computer Assisted Intervention, 2001, p. 1058.
    28. 28)
      • 34. Comaniciu, D., Meer, P.: ‘Mean shift: a robust approach toward feature space analysis’, IEEE Trans. Pattern Anal. Mach. Intell., 2002, 24, (5), pp. 603–619.
    29. 29)
      • 35. Unnikrishnan, R., Pantofaru, C., Hebert, M.: ‘Toward objective evaluation of image segmentation algorithms’, IEEE Trans. Pattern Anal. Mach. Intell., 2007, 29, pp. 929–944.
    30. 30)
      • 26. Topkaya, I.S., Erdogan, H., Porikli, F.: ‘Tracklet clustering for robust multiple object tracking using distance dependent Chinese restaurant processes’, Signal Image Video Process., 2016, 10, (5), pp. 795–802.
    31. 31)
      • 18. Antoniak, E.C.: ‘Mixtures of Dirichlet processes with applications to Bayesian nonparametric problems’, Ann. Stat., 1974, 2, (6), pp. 1152–1174.
    32. 32)
      • 23. Blackwell, D., MacQueen, J.B.: ‘Ferguson distributions via Polya urn schemes’, Ann. Stat., 1973, 1, (2), pp. 353–355.
    33. 33)
      • 32. Cliord, P.: ‘Markov random fields in statistics’, in Grimmett, G., Welsh, D. (Eds.): ‘Disorder in physical systems: a volume in honour of John M. Hammersley’ (Oxford University Press, 1990), pp. 19–32.
    34. 34)
      • 3. Turek, M.W., Freedman, D.: ‘Multiscale modeling and constraints for maxow/min-cut problems in computer vision’, Conference on Computer Vision and Pattern Recognition Workshop (CVPRW'06), 2006, p. 180.
    35. 35)
      • 9. Ahmadvand, A., Daliri, M.R.: ‘Brain MR image segmentation methods and applications’, OMICS J. Radiol., 2014, 3, p. e130, doi: 10.4172/2167-7964.1000e130.
    36. 36)
      • 19. Sudderth, B.E., Jordan, I.M.: ‘Shared segmentation of natural scenes using dependent Pitman–Yor processes’, Adv. Neural Inf. Process. Syst., 2008, pp. 1585–1592.
    37. 37)
      • 24. Blei, D.M., Frazier, P.I.: ‘Distance dependent Chinese restaurant processes’, J. Mach. Learn. Res., 2011, 12, pp. 2461–2488.
    38. 38)
      • 39. Freixenet, J., Muñoz, X., Raba, D., et al: ‘Yet another survey on image segmentation: region and boundary information integration’. Computer Vision – ECCV 2002, Copenhagen, Denmark, May 2002, pp. 408–422.
    39. 39)
      • 22. Chatzis, S.P.: ‘A Markov random field-regulated Pitman–Yor process prior for spatially-constrained data clustering’, Pattern Recognit., 2013, 46, (6), pp. 1595–1603.
http://iet.metastore.ingenta.com/content/journals/10.1049/iet-cvi.2016.0429
Loading

Related content

content/journals/10.1049/iet-cvi.2016.0429
pub_keyword,iet_inspecKeyword,pub_concept
6
6
Loading
Print this page
This is a required field
Please enter a valid email address