架構於小波關聯隱藏馬可夫樹模式的｜國立中央大學博碩士論文系統

簡易檢索 / 詳目顯示

回結果列表

研究生：	陳怡萍 Yi-Ping Chen
論文名稱：	架構於小波關聯隱藏馬可夫樹模式的 Texture Image Segmentation based onWavelet Contextual Hidden Markov Tree Models
指導教授：	曾定章 Din-chang Tseng
口試委員:
學位類別：	碩士 Master
系所名稱：	資訊電機學院 - 資訊工程學系 Department of Computer Science & Information Engineering
畢業學年度：	95
語文別：	英文
論文頁數：	86
相關次數：	點閱：9 下載：0
分享至:	分享至facebook 分享至twitter

查詢本校圖書館目錄查詢臺灣博碩士論文知識加值系統勘誤回報

在本論文中，我們提出了關聯隱藏馬可夫樹模式 (contextual hidden
Markov tree model, CHMT) 和邊界精細化 (boundary refinement) 方法來
做紋理影像分割。關聯隱藏馬可夫樹模式是由建立在小波轉換架構下的
隱藏馬可夫樹模式 (hidden Markov tree model, HMT) 改良而來的。隱藏
馬可夫樹模式是用來捕捉小波係數之統計特性的一種樹狀結構機率模
式；隱藏馬可夫樹可以完整的描述小波係數的繼承性 (persistence
property)，但不太具有聚集性 (clustering property)。而關聯隱藏馬可夫樹
模式則使用擴增點 (extended nodes) 的觀念，來加強隱藏馬可夫樹模式的
聚集性。
在影像分割的應用上，因為邊界精細化方法加入影像像素的位置資
訊，區分為同質區域及邊界區域，因此，我們使用邊界精細化方法來加
強粗分割的正確性。首先，對於每一種紋理影像，利用關聯隱藏馬可夫
樹模式訓練一組代表此紋理影像的關係參數；接著利用這些參數算出不
同解析度區塊的最大相似度函數值做第一次分割；但分割結果，解析度
愈高的影像正確率愈低。接著依照影像的區域性融合不同解析度的分割
結果以得到更精確的分割結果。

A multiscale texture image segmentation approach based on the
contextual hidden Markov tree (CHMT) model and boundary refinement is
proposed. The hidden Markov tree models (HMT) is a statistical model of tree
structure for capturing properties of wavelet coefficients. The HMT model
describes persistence property of wavelet coefficients, but loses clustering
property. We have proposed the CHMT model which improved from the HMT
model by enhancing the clustering property.
The CHMT model reinforces clustering property by using extended
coefficients without changing the wavelet tree structure; thus the HMT
training scheme can be easily modified to estimate the parameters of the
CHMT model.
In this study, the CHMT model is applied for texture segmentation. For
each texture, we use the CHMT model to train a set of parameters and then
utilize these parameters compute likelihood functions for all mulitscale
squares of a test image. At last, we segment the test image with the principle
of maximum likelihood. Only based on the CHMT model, the segmentation
results are not good enough when the size of dyadic square is small; thus the
boundary refinement algorithm is adopted to fuse the multiscale square to get
better-quality segmented results. The segmented results based on the HMT
and CHMT models are compared to show the improvement of the CHMT
model over the HMT model; moreover, the boundary refinement algorithm is
also evaluated to show its ability.

摘 要 .................................................... II
誌 謝 ....................................................III
目 錄 ................................................... IV
第一章 緒論................................................一
第二章 相關研究............................................二
第三章 小波域的機率模式....................................三
第四章 粗分割後的融合方法.................................. 四
第五章 多重解析度影像分割..................................五
第六章 實驗與討論..........................................六
第七章 結論................................................七
英文版論文..................................................八
Abstract ............................................................................................................ ii
Contents ........................................................................................................... iii
List of Figures ................................................................................................... v
List of Tables .................................................................................................... ix
Chapter 1 Introduction ...................................................................................... 1
1.1 Motivation ........................................................................................ 1
1.2 System overview .............................................................................. 2
1.3 Thesis organization .......................................................................... 3
Chapter 2 Related Works ................................................................................... 5
2.1 Model-based image segmentation .................................................... 5
2.2 HMM-based image segmentation .................................................... 6
2.3 HMT-based image segmentation ...................................................... 7
Chapter 3 Statistical Image Models ................................................................ 10
3.1 Gaussian Mixture Models .............................................................. 10
3.2 The wavelet transform ................................................................... 11
3.3 Probabilistic model for a single wavelet coefficient ...................... 14
3.4 Probabilistic models based on wavelet transforms ........................ 16
3.4.1 Hidden Markov tree models ............................................... 19
3.4.2 Contextual hidden Markov tree model ............................... 20
3.5 Wavelet domain CHMT model training ......................................... 23
Chapter 4 Fusion of the Raw Segmented Results ........................................... 31
4.1 Context-based interscale fusion ..................................................... 32
4.1.1 Interscale fusion concepts ................................................... 32
4.1.2 Interscale fusion algorithm ............................................... 335
4.2 Boundary refinement ...................................................................... 37
Chapter 5 Multiscale Segmentation Using Contextual Hidden Markov Tree
Models ............................................................................................ 39
5.1 Multiscale image segmentation framework ................................... 39
5.2 Multiscale segmentation ................................................................ 40
5.3 Pixel-level segmentation ................................................................ 42
5.3.1 Spatial-domain pixel miture Gaussian model training ....... 42
5.3.2 Pixel segmentation .............................................................. 43
5.4 Boundary refinement ...................................................................... 43
Chapter 6 Experiment and Discussions .......................................................... 47
6.1 Experimental results ....................................................................... 47
6.2 Discussions ..................................................................................... 58
Chapter 7 Conclusions and Future Works ....................................................... 60
References ....................................................................................................... 61

                                

[1] Bouman, C. and M. Shapiro, “A multiscale random field model for
Bayesian image segmentation,” IEEE Trans. Image Processing, vol.38,
no.2, pp.162-177, 1994.
[2] Charles, F. V. L, Introduction to Scientific Computing: A Matrix Vector
Approach Using MATLAB, Prentice Hall, New York, 1996.
[3] Chen, J.-L. and A. Kundu, “Automatic unsupervised texture
segmentation using hidden Markov model,” in Proc. IEEE Int. Conf.
Acoustics, Speech and Signal Processing, Minneapolis, Minnesota, Apr.
27-30, 1993, pp.21-24.
[4] Choi, H. and R. G. Baraniuk, “Multiscale texture segmentation using
wavelet-domain hidden Markov models,” in Proc. 32th Asilomar Conf.
Signals, System and Computers, Pacific Grove, CA, Nov.1-4, 1998,
pp.1692-1697.
[5] Choi, H. and R. G. Baraniuk, “Image segmentation using wavelet-domain
classification,” in Proc. SPIE Conf. Mathematical Modeling, Bayesian
Estimation, and Inverse Problems, Denver, Colorado, July 21-23, 1999,
pp.306-320.
[6] Choi, H. and R. G. Baraniuk, “Multiscale image segmentation using
wavelet-domain hidden Markov models,” IEEE Trans. Image Processing,
vol.10, no.9, pp.1039-1321, 2001.
[7] Crouse, M. S., R. D. Nowak, and R. G. Baraniuk, “Wavelet-based
statistical signal processing using hidden Markov models,” IEEE Trans.
Signal Processing, vol.46, no.4, pp.886-902, 1998.
[8] Etai, M. and M. Aladjem, “Boundary refinements for wavelet-domain
multiscale texture segmentation,” Image and Vision Computing, vol.23,no.13, pp.1150-1158, 2005.
[9] Fan, G. and X. G. Xia, “Wavelet-based statistical image processing using
hidden Markov tree model,” in Proc. 34th Annual Conf. Information
Sciences and Systems. Princeton, New Jersey, Mar.15-17, 2000.
[10] Fan, G. and X. G. Xia, “Maximum likelihood texture analysis and
classification using wavelet-domain hidden Markov model,” in Proc.
34th Asilomar Conf. Signals, System and Computers, Pacific Grove, CA,
Oct.29-Nov.1, 2000, pp.921-925.
[11] Fan, G. and X. G. Xia, “On context-based bayesian image segmentation:
joint multi-context and multiscale approach and wavelet-domain hidden
Markov models,” in proc. 35th Asilomar Conf. Signals, System and
Computers, Pacific Grove, CA, Nov.4-7, 2001, pp.1146-1150.
[12] Fan, G. and X. G. Xia, “Wavelet-based texture analysis and synthesis
using hidden Markov models,” IEEE Trans. Fundamental Theory and
Applications, vol.50, no.1, pp.106-120, 2003.
[13] Fan, Y., T. Jiang, and D. J. Evans, “Volumetric segmentation of brain
images using parallel genetic algorithm,” IEEE Trans. Medical Imaging,
vol.21, no.8, pp.904-909, 2002.
[14] Ginneken, B., A. F. Frangi, J. J. Staal, B. M. Romeny, and M. A.
Viergever, “Active shape model segmentation with optimal features,”
IEEE Trans. Medical Imaging, vol.21, no.8, pp.924-933, 2002.
[15] Kaplan, L. M., “Extended fractal analysis for texture classification and
segmentation,” IEEE Trans. Image Processing, vol.8, no.11, pp.
1572-1585, 1999.
[16] Krishnamachari, S. and R. Chellappa, “Multiresolution Gauss-Markov
random field models for texture segmentation,” IEEE Trans. ImageProcessing, vol.6, no.2, pp.251-267, 1997.
[17] Lerman, J., S. Kulkarni, and J. Koplowitz, “Multiresolution chain coding
of contours,” in Proc. IEEE Int. Conf. Image Processing, Princeton, New
Jersey, Nov.13-16, 1994, pp.615-619.
[18] Li, J., A. Najmi, and R. M. Gray, “Image classification by a
two-dimensional hidden Markov model,” IEEE Trans. Acoustics, Speech
and Signal Processing, vol.48, no.2, pp.517-533, 2000.
[19] Ling, P., Z. M. Zhong, and J. L. Ma, “Texture image segmentation based
on wavelet-domain hidden Markov models,” in Proc. IEEE Int. Conf.
Geoscience and Remote Sensing Symposium, Anchorage, Alaska,
Sep.20-24, 2004, pp.3829-3832.
[20] Marroquin, J. L., B. C. Vemuri, S. Botello, F. Calderon, and A.
Fernandez-Bouzas, “An accurate and efficient Bayesian method for
automatic segmentation of brain MRI,” IEEE Trans. Medical imaging,
vol.21, no.8, pp.934-945, 2002.
[21] Pesquet, J. C., H. Krim, and E. Hamman, “Bayesian approach to best
basis selection,” in Proc. IEEE Int. Conf. Acoust., Speech, Signal
Process. ICASSP, Atlanta, GA, 1996, pp.2634-2637.
[22] Pitiot, A., A. W. Toga, and P. M. Thompson, “Adaptive elastic
segmentation of brain MRI via shape-model-guided evolutionary
programming,” IEEE Trans. Medical Imaging, vol.21, no.8, pp.910-923,
2002.
[23] Rabiner, L. R., “A tutorial on hidden Markov models and selected
applications in speech recognition,” IEEE Trans. Digital Object
Identifier, vol.77, no.2, pp.257-285, Feb. 1989.
[24] Ramesh, N., J. Romberg, H. Choi, R. Riedi and R. Baraniuk,“Multiscale image segmentation using joint texture and shape analysis,”
in Proc. SPIE Conf. Wavelet Applications in Signal and Image
Processing, San Diego, CA, July 30-Aug.4, 2000, pp.215-218.
[25] Shapiro, J. M., “Embedded image coding using zerotrees of wavelet
coefficients,” IEEE Trans. Signal Processing, vol.41, no.12,
pp.3445-3462, 1993.
[26] Stollnitz, E. J., T. D. DeRose, and D. H. Salesin, Wavelets for Computer
Graphic, Morgan Kaufmann, San Francisco, 1996.
[27] Tseng, D.-C. and M. Shih, “Wavelet-based image denoising using
contextual hidden Markov tree model,” Joural of Electronic Imaging,
vol.14, no.3, pp.1-12, July 2005.

簡易檢索 / 詳目顯示

相關論文