邊界偵測是在傳統的影像處理課題中，一項非常重要的應用，在大自然裡，我們也有可能需要面對這樣的問題，例如說要如何利用無線感測器快速地偵測外海汙油擴散的確實範圍。在分散式無線感測網路的環境裡，我們利用巢狀類神經網路的技術來達成邊界偵測的目的。使用這種方法，可以讓每一個感測器只需接收鄰近感測器的資訊，就可以判斷自己是否為邊界感測器。
首先，我們提出了一種演算法。考慮感測器在不同的通訊能力之下，偵測邊界的表現，決定出效能最好的感測器其通訊範圍，並模擬在不同的傳輸品質時，所能得到的結果。
其次，我們探討了另外二種同樣應用在分散式無線感測網路的邊界偵測方法，一個是基於統計逼近的統計方法，另一個是基於濾波逼近的影像處理方法。將利用這二種方法與使用巢狀類神經網路的方法所得的邊界偵測模擬結果互相比較，我們發現在低訊雜比的環境中，使用巢狀類神經網路的技術所得的模擬結果，會優於使用基於統計逼近的統計方法以及基於濾波逼近的影像處理方法所得到的結果。
第三，我們進一步探討結合巢狀類神經網路和無線感測網路的數學模型，利用數值方法求出邊界偵測的近似分析結果：感測率及誤偵率。
最後，我們比較這個使用巢狀類神經網路技術來達成無線感測網路技術的分析結果與模擬結果其感測率及誤偵率。

Edge detection is a very important application in the conventional image processing area. In real life we may need to face this critical technique, for example, understanding how to use wireless sensors to detect the correct extended range of greasiness on the sea. In the distributed wireless sensor network environment, we utilize the technique of Cellular Neural Network to reach the goal of edge detection. Using this technique, we can make each sensor only receive the information of adjacent sensors to identify itself as edge sensor or not.
First of all, we propose an algorithm, that is, according to the detecting edge results of sensors under different communication abilities, we determine the most efficient communication range of sensors and make the simulations in different communication qualities.
Moreover, we discuss the other two edge detection methods which are also used in the distributed wireless sensor network: one is the statistical-based approach and the other is the filter-based approach applied in the image processing. Comparing the edge detecting simulation results of this two methods and the method of Cellular Neural Network, we find that in the low SNR environment, the results of Cellular Neural Network we gained might be better than the ones of the other two.
Third, we discuss in advance the mathematical model combined by the Cellular Neural Network and wireless sensor network, and then, utilize the numerical recipes to obtain the edge detection's approximate analytical results, the probability of detection, and the probability of false alarm.
Finally, we compare the analytical results' probability of detection and false alarm with the ones of simulation results which belong to the edge detection found by the technique of the combination of wireless sensor network and Cellular Neural Network.

Bibliography
【1】 L.O.Chua, L.Yang, Cellular Neural Networks : Theory , IEEE Trans. on Circuits and Systems 35 1257, 1998.
【2】 L.O.Chua, L.Yang, Cellular Neural Networks : Applications , IEEE Trans. on Circuits and Systems 35 1273, 1998.
【3】 Cs. Rekeczky, "MATCNN - Analogic Simulation Toolbox for Matlab", Version 1.0, DNS-11-1997, Technical Report, Analogical and Neural Computing Laboratory, Computer and Automation Institute of the Hungarian Academy of Sciences, Budapest, September 1997.
【4】 VisMouse - SimCNN , Visual Mouse Platform and Multi-Layer CNN Simulator, User's Guide", Analogical and Neural Computing Laboratory, Hungarian Academy of Sciences, Budapest, 1999.
【5】 Localized Edge Detection in a Sensor Field, Krishna Kant Chintalapudi, Ramesh Govindan. University of Southern California, Los Angeles, California, USA, 90007.
【6】 Pei-Kai Liao, Min-Kuan Chang and C.-C. Jay Kuo, "Distributed Edge Detection with Composite Hypothesis Test in Wireless Sensor Networks," in Proc. IEEE Global Telecommunication Conference (Globecom) 2004.
【7】 Pei-Kai Liao, Min-Kuan Chang and C.-C. Jay Kuo, "A Statistical Approach to Distributed Edge Detection in Wireless Sensor Networks," in Proc. SPIE Defense and Security Symposium 2004.
【8】 Pei-Kai Liao, Min-Kuan Chang and C.-C. Jay Kuo, "Distributed Edge Sensor Detection with One- and Two-Level Decisions," in Proc. IEEE International Conference on Acoustic, Speech and Signal Processing (ICASSP) 2004.
【9】 Gang Liu,Robert M.Haralick,“Assignment Problem in Edge. Detection Performance Evaluation ”, Proceedings. IEEE Conference on Computer Vision and Pattern. Recognition, volume: 1, 2000, pp. 26-31.
【10】 Akyildiz LF, Su WL, Sankarasubramaniam Y, Cayirci E. A survey on sensor networks. IEEE Communications Magazine, 2002, 40(8):102~114.
【11】 Sung Park, Andreas Savvides, Mani B. Srivastava: Simulating networks of wireless sensors. Winter Simulation Conference 2001: 1330-1338.
【12】 William H. Press, Numerical recipes in C :the art of scientific computing, Cambridge University Press,1992.New York.
【13】 Gradshtein, I. S., Ryzhik, I. M., Jeffrey, Alan., Table of integrals, series, and products., San Diego, Academic Press,c2000.
【14】 Milton Abramowitz, Stegun, Irene A., Handbook of mathematical functions :with formulas, graphs, and mathematical tables, New York, Dover Publications,1965.
【15】 Scheid, Francis J., SCHAUM'S OUTLINES NUMERICAL ANALYSIS 2/E 1988, New York :McGraw-Hill,1989.