臺灣博碩士論文加值系統

颱風是大氣循環系統的重要部分，但是颱風也是一個具有巨大破壞力的天氣系統。颱風會帶來強風、豪雨、水災和海水倒灌等可怕的災害，為了預防以及減少颱風所帶來的災害，提早預測颱風動態是一個關鍵。 颱風眼是一個用來預測颱風強度以及移動路徑非常重要的依據，因此有許多學者提出了從衛星影像中進行颱風中心的定位、颱風強度的估計和移動路徑預測的研究成果。然而，在進行颱風中心定位之前有一個非常重要的步驟是從衛星圖像中切割出颱風雲層。大部分從衛星影像中定位颱風的理論是以颱風樣板比對的方式達成，但是樣板比對方法往往需要複雜的電腦運算。在這篇文章中，我們提出了一個新的颱風定位方法，利用分割紅外線衛星影像為數個薄片（slices），並且使用形態學運算和影像分類法自動定位颱風。進一步，我們也使用颱風衛星影像的薄片去估計颱風中心位置，並且針對近年在台灣發生的颱風利用所提出的定位方法去追蹤颱風中心。為了評估所提出的颱風定位方法，我們運用了台灣中央氣象局所提供從1995年到2005年之間的64個颱風的紅外線衛星影像進行實驗，每個颱風包括成熟期到衰敗期之衛星圖像。根據實驗結果，我們提出的颱風特徵有助於準確定位颱風以及預估颱風中心，颱風定位的測試結果也符合我們的期望。

Typhoons are an important part of the atmospheric circulation system. Typhoons also are weather systems with vast destructive power. Typhoon inflicts terrible damage due to thunderstorms, violent winds, torrential rain, flooding and extremely high tides. Improving the early typhoon forecast capability is a key to the disaster prevention. The typhoon center is a very important indicator of in typhoon central power estimation and moving path forecast. Many scholars made efforts in typhoon center location, typhoon intensity estimation and moving path prediction from the satellite images. However, one of the most important processing of the typhoon center location is to segment the typhoon cloud from the satellite images. Many important techniques for locating typhoon based on satellite images were almost developed using typhoon pattern matching methods. The pattern matching scheme under the above architecture is inherently complex in computation. In this thesis, we proposed a novel method that can partition the infrared satellite image into slices and locate the typhoon automatically using morphology operations and image classification techniques. In addition, we use the slices of typhoon satellite image to estimate the typhoon center. We applied our approach to locate and track the center of different typhoons occurred in recent years and achieved high accuracy. In order to evaluate the proposed methods, the infrared satellite images of 64 typhoons from mature period to decaying period during the period between 1995 and 2005 provided by Central Weather Bureau of Taiwan are used. The typhoon features we extracted are useful to identify typhoon and estimate the typhoon center. The experimental results are also satisfied our expectation in the locating of typhoons.

ACKNOWLEDGMENTS..........................................i
ABSTRACT.................................................ii
摘要......................................................iii
TABLE OF CONTENTS........................................iv
LIST OF FIGURES..........................................v
LIST OF TABLES...........................................vi
CHAPTER 1 INTRODUCTION...................................1
1.1 Introduction.........................................1
1.2 Motivation and Objective.............................2
1.3 Thesis Organization..................................4
CHAPTER 2 BACKGROUNDS....................................5
2.1 Typhoon Definition...................................5
2.2 Infrared Satellite Image.............................6
2.3 Review of Typhoon Location Methods...................7
2.3.1 Typhoon Identification.............................7
2.3.2 Typhoon Center Location............................10
CHAPTER 3 TYPHOON LOCATION METHODS.......................14
3.1 Image Luminance Scaling..............................15
3.2 Thin Line Elimination and Image Segmentation.........18
3.2.1 Image Binarization.................................19
3.2.2 Image Erosion......................................20
3.2.3 Region Growing.....................................22
3.2.4 Typhoon Candidate Segmentation.....................24
3.3 Image Slicing........................................26
3.4 Typhoon Recognition..................................27
3.5 Typhoon Center Location..............................31
CHAPTER 4 EXPERIMENTAIL RESULTS..........................32
4.1 Typhoon Satellite Image Database.....................32
4.2 Experimental Results of Typhoon Location.............32
CHAPTER 5 CONCLUSION AND FUTURE WORK.....................40
5.1 Conclusion...........................................40
5.2 Future Work..........................................42
REFERENCE................................................43

[1]http://en.wikipedia.org/wiki/Hurricane_Katrina, May 22, 2006.
[2]V.F. Dvorak, “Tropical Cyclone Intensity Analysis and Forecasting from Satellite Imagery,” Monthly Weather Review, Vol. 103, No. 5, May 1975, pp. 420–430.
[3]C.L. Yip and K.Y. Wong, “Efficient and Effective Tropical Cyclone Eye Fix using Genetic Algorithms,” Springer-Verlag Berlin Heidelberg, 2004, pp. 654-660.
[4]S.T. Lee and N.K. Liu, “Tropical Cyclone Identification and Tracking System using Integrated Neural Oscillatory Elastic Graph Matching and Hybrid RBF Network Track Mining Techniques,” IEEE Transactions on Neural Networks, Vol. 11, Issue 3, May 2000, pp. 680-689.
[5]S.T. Lee and N.K. Liu, “An Elastic Contour Matching Model for Tropical Cyclone Pattern Recognition,” IEEE International Conference on Systems, Man, and Cybernetics, Vol. 31, 2001, pp. 413-417.
[6]M.I. Hossain, J. Liu and J. You, "Tropical Cyclone Pattern Recognition for Intensity and Forecasting Analysis from Satellite Imagery," IEEE International Conference on Systems, Man, and Cybernetics, Vol. 6, 12-15 Oct 1999, pp. 851-856.
[7]C.L. Yip and K.Y. Wong, “An Intelligent Tropical Cyclone Eye Fix System using Motion Field Analysis,” IEEE International Conference on Tools with Artificial Intelligence, 2005, pp. 652-656.
[8]Q.P. Zhang, L.L. Lai, and W.C. Sun, “Intelligent Location of Tropical Cyclone Center,” Proceedings of the Fourth International Conference on Machine Learning and Cybernetics, Vol. 1, 2005, pp. 423-428.
[9]P. Wang, P.L Yang, W.L Li and H.Q. Lu, “Extracting the Rotation Feature of the Developing Typhoon,” Proceedings of International Conference on Machine Learning and Cybernetics, Vol. 8, 18-21 Aug 2005, pp. 5229-5234.
[10]L.D Stefano. Mattoccia and M. Mola, “An Efficient Algorithm for Exhaustive Template Matching based on Normalized Cross Correlation,” Proceedings of 12th International Conference on Image Analysis and Processing, 2003, pp. 322-327.
[11]B. Jie, W. Hongqing and T. Zuyu,“The Deriving of Cloud Motion Winds from IR Images of GMS,” Acta Scientiarum Naturalium Universitatis Pekinensis, Vol. 33, No.1, 1997, pp. 85-92.
[12]K.A. Emanuel, “The Theory of Hurricanes,” Annu. Rev. Fluid Mech., 23, 1991, pp. 179-196.
[13]J.J. Park, K.A. Park, K. Kim and Y.Y. Youn, “Statistical Analysis of Upper Ocean Temperature Response to Typhoons from ARGO floats and Satellite data,” Proceedings of International Conference on Geoscience and Remote Sensing Symposium, Vol. 4, 25-29 July 2005, pp. 2564 – 2567.
[14]C.H. Poon, C.P. Chau and M. Ghadiali, “Using Fuzzy Mathematical Morphology for Locating Tropical Cyclones in Satellite Imagery,” IEEE International Symposium on Circuits and Systems, Vol. 2, 1997, pp. 1353-1356.
[15]Z. Liu, B. Wu and J. Xue, “The Non-eye Typhoon Center Locating with Parallel Computing,” Proceedings of the Fourth International Conference on Parallel and Distributed Computing, Applications and Technologies, 2003, pp. 488-491.