Abstract
The wavelet-domain hidden Markov tree (HMT) model provides a powerful approach for image modeling and processing because it captures the key features of the wavelet coefficients of real-world data. However, it is usually assumed that the subbands at the same level are independent in the traditional HMT model. This paper proposes a modified HMT model, SHMT-S, in which a vector constructed from the coefficients at the same location of the subbands of the same level, is controlled by a hidden state. Meanwhile we also use the vector Laplace mixture distributions to fit the wavelet coefficients vector, which is peakier in the center and has heavier tails compared with Gaussian distribution. By using the HMT segmentation framework, we develop SHMT-S based segmentation methods for image textures and dynamic textures. The experimental results demonstrate the effectiveness of the proposed method.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Chen J, Zhao G-Y, Salo M, Rahtu E, Matti P (2013) Automatic dynamic texture segmentation using local descriptors and optical flow. IEEE Trans Image Process 22:326–339
Crouse MS, Nowak RD, Baraniuk RG (1998) Wavelet-based statistical signal processing using hidden Markov models. IEEE Trans Signal Process 46:886–902
Fan Guo-Liang, Xia Xiang-Gen (2003) Wavelet-based texture analysis and synthesis using hidden Markov models. IEEE Trans Circ Syst 50:106–120
Romberg JK, Choi H, Baraniuk RG (2001) Bayesian tree-structured image modeling using wavelet-domain hidden Markov models. IEEE Trans Image Process 10:1056–1068
Choi H, Baraniuk RG (2001) Multiscale image segmentation using wavelet-domain hidden Markov models. IEEE Trans Image Process 10:1309–1321
Durand J-B (2004) Computational methods for hidden Markov tree models—an application to wavelet trees. IEEE Trans Signal Process 52:2551–2560
Rasheed HA, Nadia JA (2014) Some Bayes’ estimators for Laplace distribution under different loss functions. J Babylon Univ 22:975–983
Jordan M, Kleinberg J, Bernhard S (2006) Pattern recognition and machine learning, pp 424–429
Brodatz P (1966) Textures: a photographic album for artists & designers, dover (1966)
Péteri R, Fazekas S, Huiskes MJ (2010) DynTex: a comprehensive database of dynamic textures. Pattern Recogn Lett 31:1627–1632
Acknowledgments
This work is partially supported by National Natural Science Foundation of China 61371175 and Fundamental Research Funds for the Central Universities HEUCFQ20150812.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2016 Springer Science+Business Media Singapore
About this paper
Cite this paper
Qiao, Y., Zhao, G. (2016). Modified Wavelet Domain Hidden Tree Model for Texture Segmentation. In: Park, J., Jin, H., Jeong, YS., Khan, M. (eds) Advanced Multimedia and Ubiquitous Engineering. Lecture Notes in Electrical Engineering, vol 393. Springer, Singapore. https://doi.org/10.1007/978-981-10-1536-6_81
Download citation
DOI: https://doi.org/10.1007/978-981-10-1536-6_81
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-10-1535-9
Online ISBN: 978-981-10-1536-6
eBook Packages: Computer ScienceComputer Science (R0)