臺灣博碩士論文加值系統

本論文延續[23]之研究成果提出一種新的影像分割方法，稱為自我組織融合演算法(SOF)，此演算法係利用改良式型態學分水嶺法所產生之影像區塊和其相對之特徵值，根據事先定義之適當的相似度測量值，用以區塊大小加權，對於鄰近區塊之特徵值做同步更新；再計算更新後之特徵值其統計參數，以得到各區塊之合併閥值。SOF演算法藉著遞迴前述之步驟－更新特徵值、同步合併，直到收斂而停止，而得到分割結果。其自我停止的特性，使SOF演算法在應用時無須事先定義物件數目；而本身可同步更新與合併的特點，大幅降低了計算時間，且深具全平行硬體實現之潛力；其依特徵值之區域統計參數，包含了區域延伸之概念，提供較客觀的合併準則。與傳統循序合併之方法做比較， SOF除了提升計算效率，亦得到精確的分割結果。

The thesis extends the work of [23] and presents a novel method called self-organizing fusion (SOF) for performing fast image segmentation. Characteristics of SOF are explored and discussed, both theoretically and empirically. The essence of SOF is that objects are extracted through alternating processes of updating and merging until convergence. Such concurrent updating creates a self-organizing fusion behavior that facilitates identification of regions comprising the same object. The method is computationally efficient as both updating and merging are conducted in parallel fashion, and since parameters selection is done for local regions, it is able to deal with fairly complex images.

CONTENTS

CHAPTER 1 INTRODUCTION 8
1-1 Review of Image Segmentation 8
1-1.1 Histogram-Based Approach 8
1-1.2 Edge-Based Approach 10
1-1.3 Region-Based Approach 11
1-1.4 The Hybrid Approach 12
1-2 Motivation of the work 12
CHAPTER 2 METHOD OF SELF-ORGANIZING FUSION 17
2-1. Feature Fusion 17
2-2. Parallel Merging and Adjacency Extension 21
2-3. Direction Map 24
CHAPTER 3 SIMULATION RESULTS 28
3-1 Applying SOF to Image Segmentation 28
3-2 Noise Tolerance 35
CHAPTER 4 CHARACTERIZATIONS OF SOF 38
4-1 Stability Analysis 38
4-2 The ��(t) Function 43
4-3 The �� parameter 44
4-4 Without the Feature Fusion 47
CHAPTER 5 CONCLUSIONS AND DISSCUSSIONS 49
REFERENCES 51

[1] P. Suetens, P. Fua, and A. J. Hanson, “Computational strategies for object recognition,” ACM Comput. Surv., vol. 24, pp. 5-61, Mar. 1992.
[2] M. Boman, K. Hohne, U. Tiede, and M. Riemer, “3-D segmentation of MR images of the head for 3-D display,” IEEE Trans. Med. Image, vol. 9, pp. 253-277, June 1990.
[3] P. Willemin, T. Reed, and M. Kunt, “Image sequence coding by split and merge,” IEEE Trans. Commun., vol. 39, pp. 1845-1855, Dec. 1991.
[4] R. C. Gonzalez and R. E. Woods, Digital Image Processing, 2nd Edition, Prentice-Hall, New Jersey 2002.
[5] N. R. Pal and S. K. Pal, “A review on image segmentation techniques,” Pattern Recognition Soc., pp. 1277-1294, 1993.
[6] J. Basak, B. Chanda, and D. D. Majumder, “On edge and line linking with connectionist models,” IEEE Trans. System, Man, and Cybernetics, vol. 24, pp. 413-428, Mar. 1994.
[7] S. A. Hojjatoleslami and J. Kittler, “Region growing: a new approach,” IEEE Trans. Image Processing, vol. 7, pp.1079-1084, Jul. 1998.
[8] S. Chen, W. Lin, and C. Chen, “Split-and-merge image segmentation based on localized feature analysis and statistical tests,” CVGIP: Graph. Models Image Processing, vol. 53, pp. 457-475, Sep. 1991.
[9] T. Pavlidis and Y. Liow, “Integrating region growing and edge detection,” IEEE Trans. Pattern Anal. Machine Intell., vol. 12, pp. 225-233, Mar. 1990.
[10] L. D. Griffin, A. C. F. Colchester, and G. P. Robinson, “Scale and segmentation of grey-level images using maximum gradient paths,” Image Vis. Comput., vol. 10, pp. 389-402, 1992.
[11] D. Hagyard, M. Razaz, and P.Atkin, “Analysis of watershed algorithms for greyscale images”, IEEE Proc. Int. Conf. Image Processing, 1996, pp. 41-44.
[12] Nock R. and Nielsen F. ‘Statistical region merging’, IEEE Trans. Pattern Anal. Machine Intell., vol. 26, issue 11, Nov. 2004, pp. 1452-1458.
[13] L. Vincent and P. Soille, “Watersheds in digital spaces: An efficient algorithm based on immersion simulations,” IEEE Trans. Pattern Anal. Machine. Intell., vol. 13, pp. 583-598, June 1991.
[14] T. Pavlidis, Structural Pattern Recognition, New York: Springer, 1980.
[15] K. Haris, S. N. Efstratiadis, N. Maglaveras, and A. K. Katsaggelos, “Hybrid image segmentation using watersheds and fast region merging,” IEEE Trans. Image Processing, vol. 7, no. 12, pp. 1684-1699, Dec. 1998.
[16] J. J. Hopfield, ‘Pattern recognition computation using action potential timing for stimulus representation’, Nature, vol. 376, 1995, pp. 33-36.
[17] J. M. Gauch, ‘Image segmentation and analysis via multiscale gradient watershed hierarchies’, IEEE Trans. Image Processing, vol. 8, no. 1, Jan. 1999, pp. 69-79.
[18] Y. W. Yu, ‘Image Segmentation based on improved watershed techniques’, Master thesis, National Taiwan Ocean University, Jun 2000.
[19] A. Papoulis, and S.U. Pillai, , Probability, Random Variables, and Stochastic Processes, 4th Ed. McGrawHill, 2002.
[20] B. Zeidman, Designing with FPGAs and CPLDs, CMP Books, Sep., 2002.
[21] S. Haykin, , Neural Networks, 2nd Ed. Prentice-Hall, 1999.
[22] T. Ban and S. Abe. ’Spatially Chunking Support Vector Clustering Algorithm,’ Proc. IEEE Intnl Joint Conf. on Neural Networks, Vol. 1, pp.25-29, July, 2004.
[23] F. W. Su, ‘Similarity Fusion Algorithm Applied to Image Segmentation’, Master thesis, National Taiwan Ocean University, Jun 2004.