臺灣博碩士論文加值系統

在這篇論文中包含三個主要的章節,
第二章. 自我組織神經網路用於震測水平層連接
第三章. 循序分類於網路節點成長之研究
第四章. 類神經網路於震測圖型之強健性辨識
在第二章中, 我們基於自我組織映射的原理設計了一個演算法來連接地震圖中的震測水平層. 神經元的拓撲是一個線性圖形用來表示一條水平層. 神經元的權重係數是被地震圖的點 (Peak Point) 以自我組織方式來調整的. 一個 Peak Point 抓取一個神經元. Moment 被用來決定新的神經元建立的位置. 我們用這個演算法用再模擬的以及真實的地震圖上面, 得到相當好的結果.
在第三章中, 我們結合了兩個重要的技術, the approximating posteriori probability functions of the classes in the outputs of the trained multilayer perceptron 以及 sequential classification technique, 我們可以從 perceptron 和 two-layer perceptron 中得到最小的節點數目. 我們將這個技術用於典型的 exclusive OR 的問題和震測圖型識別上面. 網路節點降低的比率非常好.
在第四章中, multilayer perceptron 類神經網路被訓練當作一個分類器, 用於辨識震測圖型. 在平移, 旋轉, 大小可以維持不變的 Seven moments 被運用在每一個震測圖型的特徵產生. 在這個系統裡面, 有訓練圖型組和測試圖型組. 在測試圖型組裡面包含了不同等級的雜訊. Multilayer perceptron 在初始的時候先用無雜訊的震測圖型進行訓練. 在訓練收斂之後, 神經網路用於辨識包含有雜訊的測試圖型組. 接著在訓練圖型組中加入一些雜訊較高辨識錯誤的圖型用來重新訓練. 這樣子的訓練和辨識過程反覆幾次的階段. 在每一個訓練階段, 把收斂好的網路用於辨識在 Mississippi Canyon 的真實震測資料. 結果在使用到較高的雜訊訓練階段, Bright spot 圖型可以被偵測出來.

This thesis contains three major chapters,
Chapter 2. Self-Organizing Neural Network for Seismic Horizon Linking.
Chapter 3. Node Growing of Perceptrons By Sequential Classification Technique.
Chapter 4. Neural Network for Robust Recognition of Seismic Patterns.
In Chapter 2, we design an algorithm based on self-organizing feature maps to link the seismic horizon in the seismogram. The topology of the neurons is a linear pattern that forms a horizon. The weighting coefficients of the neurons are self-organized by the peaks of the seismogram. One peak catches one neuron. Moment is used in the determination of new neuron creation position. We have applied the algorithm on simulated and real seismograms and the results are quite well.
In Chapter 3, combining the important property of the approximating a posteriori probability functions of the classes in the outputs of the trained multilayer perceptron and sequential classification technique, we can get the minimum number of nodes in perceptron and two-layer perceptron. We apply the technique to the typical exclusive OR problem and seismic pattern recognition. The reduction rate of nodes is quite good.
In Chapter 4, the multilayer perceptron neural network is trained as a classifier and is applied to the recognition of seismic patterns. Seven moments that are invariant to translation, rotation, and scale, are employed for feature generation of each seismic pattern. In the system, there are training and testing pattern sets. The testing pattern set includes different noise level. The multilayer perceptron is initially trained with the training set of noise-free seismic patterns. After convergence of the training, the network is applied to the classification of the testing set of noisy seismic patterns. Some misclassified patterns with higher noise level are added to the training set for retraining. The training and classification process is repeated through several stages. The converged network at each training stage is applied to the real seismic data at Mississippi Canyon, the bright spot pattern can be detected when the stage is using higher level noisy patterns in the training.

Chapter 1. Introduction
Chapter 2. Self-Organizing Neural Network for Seismic Horizon Linking.
Abstract 3
1. Introduction 4
2. Seismic Horizon Linking 6
3. Self-organizing Neural Network for Seismic Horizon Linking 8
3.1 Self-organizing feature maps 8
3.2 Proposed network structure 9
3.3 System of self-organizing neural network for seismic horizon linking 12
3.4 Horizon linking algorithm by self-organizing neural network 17
3.5 Flowchart of horizon linking algorithm by self-organizing neural network 19
4. Experiments 20
Experiment 1. Single step of a simple example that contains 3 peak points 22
Experiment 2. Single step of an example that contains broken points 26
Experiment 3. Single step of an example with a branch 37
Experiment 4. Single step of an example with a broken peak at critical position 52
Experiment 5. Some simple examples for horizon linking 71
Experiment 6. A Bright-Spot simulation seismogram 72
Experiment 7. a Flat-Spot simulation seismogram 73
Experiment 8. A Flat-Spot simulation seismogram 74
Experiment 9. A Sealevel-Fall simulation seismogram 75
Experiment 10. Real data experiment result at Mississippi Canyon 76
5. Conclusions 80
References 81
Chapter 3. Node Growing of Perceptrons By Sequential Classification Technique.
Abstract 4
1. Introduction 5
2. Sequential Classification by Single-layer Perceptron 8
2-1. Sequential classification with growing of input nodes for two-class problem on single-layer perceptron 9
2-2. Sequential classification with growing of input nodes for m-class problem on single-layer perceptron 14
3. Sequential Classification by Two-layer Perceptron 19
3-1. Sequential classification with growing of hidden nodes for two-class problem on two-layer perceptron 20
3-2. Sequential classification with growing of hidden nodes for m-class problem on two-layer perceptron 25
4. Ordering of Input Nodes on Single-Layer Perceptron and Hidden Nodes on Two-layer Perceptron 30
4-1. Ordering of input nodes on single-layer perceptron 30
4-2. Ordering of hidden nodes on two-layer perceptron 32
5. Experiments 34
Experiment 1. 2D and two-class problem on 2 x 2 single-layer perceptron 34
Experiment 2. 3D and two-class problem on 3 x 2 single-layer perceptron 36
Experiment 3. XOR problem on 2 x 6 x 2 two-layer perceptron 38
Experiment 4. Seismic pattern recognition on 7 x 10 x 3 two-layer perceptron 40
6. Conclusions 44
References 45
Chapter 4. Neural Network for Robust Recognition of Seismic Patterns.
Abstract 3
1. Introduction 4
2. Seven Moments for Pattern Feature 6
3. The Multilayer perceptron (MLP) System 14
4. Training and Recognition of MLP in Seismic Patterns 16
5. Principle of Robust Training 18
6. System of Seismic Robust Training and Testing 20
7. Experiment 24
References 33

References of Self-Organizing Neural Network for Seismic Horizon Linking:
[1] N. Keskes and Ph. Zaccagnino, “Image analysis techniques for seismic data,” Society of Exploration Geophysicists 52nd Ann. Int. Meet. Expanded Tech. Prog. Abstr., Dallas, TX, 1982, pp.220-221.
[2] N. Keskes, Ph. Zaccagnino, and P. Mermey, “Automatic extraction of 3-D seismic horizons,” Society of Exploration Geophysicists 53rd Ann. Int. Meet. Expanded Tech. Prog. Abstr., Las Vegas, NV, 1983, pp.557-559.
[3] G. Sibille, N. Keskes, J. M. Fontaine, and J. L. Lequeux, “Enhancement of the perception of seismic facies and sequences by image analysis techniques,” Society of Exploration Geophysicists 54th Ann. Int. Meet. Expanded Tech. Prog. Abstr., Atlanta, GA, 1984, pp.447-480.
[4] S. Y. Lu, “A string-to-string correlation algorithm for image skeletonization,” Proc. Int. Joint Conf. Pattern Recognition, Munich, Germany, October 1982, pp.178-180.
[5] Yao-Chou Cheng and Shin-Yee Lu, “The binary consistency checking scheme and its applications to seismic horizon detection,” IEEE Trans. Pattern Anal. Mach. Intell. 11, April 1989, pp.439-447.
[6] K. Y. Huang, “Branch and bound search for automatic linking process of seismic horizons,” Pattern Recognition, Vol. 23, No. 6, 1990, pp.657-667.
[7] T. Kohonen, Self-Organization and Associative Memory, 2nd Edition, Springer-Verlag, 1987.
[8] T. Kohonen, Self-Organization and Associative Memory, 3rd Edition, Springer-Verlag, 1989.
[9] Kohonen, T., “The 'neural' phonetic typewriter,” Computer, Vol. 21. Issue 3, March 1988, pp.11—22.
[10] Angéniol, B., Vaubois, G., and Texier, J., “Self-organizing feature maps and the traveling salesman problem,” Neural Networks, Vol.1, 1988, pp.289-293.
[11] M. B. Dobrin, Introduction to Geophysical Prospecting, 3rd ed. New York: McGraw-Hill, 1976, ch. 10.
[12] C. E. Payton, Ed., “Seismic stratigraphy — applications to hydrocarbon exploration,” AAPG Memoir, 26, Tulsa, Am. Assn. Petroleum Geologists, 1977.
[13] K. Y. Huang, K. S. Fu, S. W. Cheng and T. H. Sheen, “Image processing of seismogram: (A) Hough transformation for the detection of seismic patterns; (B) Thinning processing in the seismogram,” Pattern recognition, vol. 18, 1985, pp.429-440.
[14] K. Y. Huang and K. S. Fu, “Detection of bright spots in seismic signal using tree classifiers,” Geoexploration 23, 1984/85, pp.121-145.
[15] Richard P. Lippman, “An introduction to computing with neural nets,” IEEE ASSP Mag., April 1987, pp.18-20.
[16] T. Kohonen, Self-Organizing Maps, Springer-Verlag, 1995.
References of Node Growing of Perceptrons By Sequential Classification Technique:
[1] D. E. Rumelhart, G. E. Hinton and R. J. Williams, "Learning internal representations by error propagation," in Parallel Distributed Processing: Expolorations in the Microstructure of Cognition. Cambridge, MA: MIT Press, ch. 8, vol. 1: Foundations, pp. 318-362, 1986.
[2] R. P. Lippmann, "An introduction to computing with neural nets," IEEE ASSP Magazine, vol. 4, pp. 4-22, April 1987.
[3] K. Funahashi, "On the approximate realization of continuous mappings by neural networks," Neural Network, vol. 2, pp. 183-192, 1989.
[4] R. Hecht-Nielson, "Theory of the back-propagation neural network," in Proc. IEEE Int. Joint Conf. Neural Networks, Washington, D. C., vol. 1, pp. 593-606, 1989.
[5] K. Hornik, M. Stinchcombe, and H. White, "Multilayer feedforward networks are universal approximators," Neural Network, vol. 2, pp. 359-368, 1989.
[6] E. D. Karnin, "A simple procedure for pruning back-propagation trained neural networks," IEEE Transactions on Neural Networks, vol. 1, no. 2, pp.239-242, 1990.
[7] J. Sietsma and R. J. F. Dow, "Creating artificial neural networks that generalize," Neural Networks, vol. 4, pp. 67-79, 1991.
[8] K. S. Fu, Sequential Methods in Pattern Recognition and Machine Learning, Academic Press, New York, 1968.
[9] D. W. Ruck, S. K. Rogers, M. Kabrisky, M. E. Oxley, and B. W. Suter, "The multilayer perceptron as an approximation to a Bayes optimal discriminant function," IEEE Transactions on Neural Networks, vol. 1, no. 4, pp. 296-298, 1990.
[10] M. B. Dobrin and C. H. Savit, Introduction to Geophysical Prospecting, Fourth Ed., McGraw-Hill, Inc., 1988.
[11] Kou-Yuan Huang, "Pattern recognition to seismic exploration," in Automated Pattern Analysis in Petroleum Exploration, edited by Ibrahim Palaz and Sailes K. Sengupta, Springer-Verlag, New York, pp. 121-154, 1992.
[12] M. K. Hu, "Visual patterns recognition by moment invariants," IRE Trans. Information Theory, vol. IT-8, pp. 179-187, 1962.
[13] G. Mirchandani and W. Cao, "On hidden nodes for neural nets," IEEE Transactions on Circuits and Systems, Vol. 36, No. 5, pp. 661-664, May 1989.
References of Neural Network for Robust Recognition of Seismic Patterns:
[1] F. Aminzadeh, S. Katz, and K. Aki, ”Adaptive neural nets for generation of artificial earthquake precursors,” IEEE Trans. Geosci. Remote Sensing, GE-32, pp. 1139-1143, 1994
[2] L. Bruzzone and D. Prieto, “A technique for the selection of kernel-function parameters in RBF neural networks for classification of remote sensing images,” IEEE Trans. Geosci. Remote Sensing, vol. 37, no.2, pp. 1179-1184, March 1999.
[3] C. H. Chen and b. Shrestha, “Classification of multi-sensor remote sensing images using self-organizing feature maps and radial basis function networks,” International Geoscience and Remote Sensing symposium, July 24-28, 2000, Proceedings, pp.711-713.
[4] P. Dysart and J. Pulli, “Regional seismic event classification at the NORESS array: Seismological measurements and the use of trained neural networks,” Bull. Of seis. Soc. of am., 80, pp. 1910-1933, 1990.
[5] K. Y. Huang, W. H. Liu, and I. C. Chang, “Hopfield model of neural networks for detection of bright-spot,” 59th Ann. Internat. Mtg., Soc, Explor. Geophys., Expanded Abstracts, 1989, pp. 444-446.
[6] K. Y. Huang and J. Y. Liaw, “Neocognitron of a neural network for seismic pattern recognition,” 62nd Ann. Internat. Meg., Soc. Explor. Geophys., Expanded Abstracts, 1992, pp. 26-29.
[7] K. Y. Huang, “Neural networks for seismic principal components analysis,” IEEE Trans. Geosci. Remote Sensing, vol. 37, pp. 297-311, Jan. 1999.
[8] D. H. Johnston, “seismic attribute calibration using neural networks,” 63rd Ann. Internat. Mtg., Soc, Explor. Geophys., Expanded Abstracts, 1993, pp. 250-253.
[9] M. D. McCormack, D. E. Zaucha, and D. W. Dushek, “First-break refraction event picking and seismic data trace editing using neural networks,” Geophysics, vol. 58, pp. 67-78, 1993.
[10] M. M. Poulton, B. K. Sternberg, and C. E. Glass, “Location of subsurface targets in geophysical data using neural networks,” Geophysics, vol. 57, pp. 1534-1544, 1992.
[11] J. Schmidt and F. A. Hadsell, “Neural network stacking velocity picking,” 62nd Ann. Internat. Mtg., Soc. Explor. Geophys., Expanded Abstracts, 1992, pp. 18-21.
[12] S. B. Serpico and F. Roli, ”Classification of multi-sensor remote sensing images by structured neural networks,” IEEE Trans, Geosci. Remote Sensing, vol. 33, no. 3, pp. 562-578, May 1995.
[13] L. X. Wang and J. M. Mendel, “Adaptive minimum prediction-error deconvolution and source wavelet estimation using Hopfield neural networks,” Geophysics, vol. 57, pp. 670-679, 1992.
[14] L. Gupta, M. R. Sayeh and r. Tammana, “A neural network approach to robust shape classification,” Pattern Recognition, vol. 23, pp. 563-568, 1990.
[15] K. Y. Huang, H. T. Yen, and C. S. Han, “Neural networks for robust recognition of printed Chinese characters,” Computer Processing of Oriental Languages, vol. 10, pp. 425-442, 1997.
[16] D. E. Rumelhart, G. E. Hinton, and R. J. Williams, “Learning internal representations by error propagation,” in Rumelhart, D. E., and McClelland, J. L., Eds., PDP, Vol. 1: M.I.T. Press, 1986, pp. 318-362.
[17] M. B. Dobrin, Introduction to Geophysical Prospecting, 3rd ed. New York: McGraw-Hill, 1976, ch. 10.
[18] C. E. Payton, Ed., “Seismic stratigraphy — applications to hydrocarbon exploration,” AAPG Memoir, 26, Tulsa, Am. Assn. Petroleum Geologists, 1977.
[19] K. Y. Huang, K. S. Fu, S. W. Cheng, and T. H. Sheen, “Image processing of seismogram: (A) Hough transformation for the detection of seismic patterns (B) Thinning processing in the seismogram,” Pattern Recognition, vol. 18, no. 6, pp. 429-440, 1985.
[20] M. K. Hu, “Visual pattern recognition by moment invariants,” IRE Trans. Info. Theory, vol. IT-8, pp. 179-187, 1962.
[21] R. C. Gonzalez and R. E. Woods, Digital Image Processing. Reading, Massachusetts: Addison-Wesley, 1993.
[22] S. Y. Kung, J. N. Hwang, and S. W. Sun, “Efficient modeling for multilayer feedforward neural nets,” Proc. IEEE Conf. On Acoustic, Speech Signal Processing, New York, 1988, pp. 2160-2163.
[23] G. Mirchandini and W. Cao, “On hidden nodes in neural nets,” IEEE Trans. Circuits and Systems, vol. 36, pp. 661-664, 1989.