車牌辨識系統廣泛應用於交通流量的監督和監看，像是贓車協尋、停車場的控管和交通流量資訊等方面。任何的辨識系統都需考量辨識的正確性和即時性，特別是針對正確性。我們所提出的車牌辨識系統使用個人數位助理(PDA)為平台並直接由PDA外掛的CMOS相機擷取影像。再由PDA進行相關處理並回傳辨識結果。一般而言，我們的系統整合了使用者的移動性並為PDA提供了一種新型態的應用。車牌辨識系統需完成車牌位置偵測、車牌文字的切割和辨識等三個主要的工作。透過影像處理技術進行車牌位置的偵測和文字的切割。倒傳遞網路則負責文字的辨識。實驗結果顯示我們所提出的系統可以有效的辨識出台灣地區大部份的車牌，包含了10個數字和26個英文字母。平均辨識率約為88%。辨識所需時間在PDA上約為3秒，在個人電腦上則小於1秒。在此篇論文中將會對車牌辨識的技術和相關模組有完整的介紹，並由實驗結果印證此架構的可行性。

Vehicle license plate recognition systems are expected to have numerous applications in traffic surveillance and monitoring, such as finding stolen cars, controlling access to parking lots and gathering traffic flow information. Any recognition system must take the accuracy and real-time response into account, especially the accuracy. Our proposed license plate recognition system uses PDA as the platform and captures the image directly from the PDA’s plug-in CMOS camera. Then the PDA directly processes the image and responds the recognition result. In general, our system integrates the mobility of user and provides a new type of application for PDA. The three main tasks of a license plate recognition system consist of detecting the license plate in the image, segmenting the license number and identifying the characters. Location and segmentation are extracted by technologies of digital image processing. The identification is done by the back propagation neural network. Experimental results show that our system can effectively recognize most Taiwanese license plates, including 10 numbers and 26 alphabet characters. The recognition rate is 88%. The recognition time takes 3 seconds on PDA and less than 1 second on personal computer. In this thesis, not only the technologies and related models used for recognizing the license plates are clearly described but also the experimental results are given to demonstrate the effectiveness of the proposed model.

誌謝 iv
ABSTRACT vi
摘要 vi
CONTENTS vi
LIST OF FIGURES x
LIST OF TABLES xiii
CHAPTER 1 INTRODUCTION 1
1.1 Motivation 1
1.2 Objectives 1
1.2 Research Restriction 2
1.3 Thesis Organization 3
CHAPTER 2 RELATED WORK 5
2.1 System Overview 5
2.2 System Architecture 6
2.3 System Analysis 7
2.4 Background Knowledge 8
2.4.1 Edge Detector 8
2.4.2 Laplacian Operator 10
2.4.3 Image Enhancement - Histogram Equalizer 11
2.4.4 Bilinear Interpolation 15
2.4.5 Binarization 16
2.4.6 Noise Reduction 17
2.5 Related Research 18
2.5.1 Character Segmentation 18
2.5.2 Character Recognition 20
CHAPTER 3 SYSTEM ARCHITECTURE 26
3.1 System Architecture Diagram 26
3.2 License Plate Detection 28
3.2.1 Preprocessing 28
3.2.2 Usage of Prewit Operator 28
3.2.3 Locating by Threshold t 29
3.3 License Number Segmentation 31
3.3.1 Image Enhancement 31
3.3.3 Laplacian Operator 33
3.3.4 Segmentation 34
3.3.5 Gradient Analysis 35
3.3.6 Error Correction 36
3.4 Recognizing the Character by Back Propagation Model 38
3.4.1 Training Stage 39
3.4.2 Testing Stage 40
CHAPTER 4 EXPERIMENTAL RESULTS AND DISCUSSION 42
4.1 Working Platform 42
4.2 Work Flow Tracing 42
4.3 Recognition Cases 46
4.4 BPN Training Model 49
4.4.1 BPN Architecture 49
4.4.2 Training Program Interface 49
4.5 Demonstration on PDA 52
4.6 Accuracy Report and Analysis 54
CHAPTER 5 CONCLUSIONS AND FUTURE WORK 56
REFERENCES 57

[1] J. Barroso, E.L. Dagless, A. Rafael and J. Bulas-Cruz, “Number plate reading using computer vision,” IEEE Int. Symp. on Industrial Electronics, vol. 3, pp.761-766, July 1997.
[2] K.K. Kim, K.I. Kim, J.B. Kim and H.J. Kim, “Learning-based approach for license plate recognition,” IEEE Signal Processing Society Workshop Neural Networks for Signal Processing X, vol. 2, no. 11-13, pp.614-623, Dec. 2000.
[3] M. Sarfraz, M.J. Ahmed and S.A.Ghazi, “Saudi Arabian license plate recognition system,” Proc. of Int. Conf. on Geometric Modeling and Graphics, pp.36-41, July 2003.
[4] C.A. Rahman, W. Badawy and A. Radmanesh, “A real time vehicle's license plate recognition system,” Proc. of IEEE Conf. on Advanced Video and Signal Based Surveillance, pp.163-166, July 2003.
[5] T. Jinshan, E. Peli and S. Acton, “Image enhancement using a contrast measure in the compressed domain,” IEEE Signal Processing Letters, vol. 10, no. 10, pp.289-292, Oct. 2003.
[6] H.C. Lui, C.M. Lee and G. Fang, “Neural network application to container number recognition,” Proc. of Computer Software and Applications Conf., pp.190-195, Nov. 1990.
[7] J.C.M. Lee, “Automatic character recognition for moving and stationary vehicles and containers in real-life images,” Int. Joint Conf. on Neural Networks, vol. 4, pp.2824-2828, July 1999.
[8] J.H. Park and V. Govindaraju, “Active character recognition using “A*-like” algorithm,” Proc. of IEEE Conf. on Computer Vision and Pattern Recognition, vol. 2, pp.82-87, June 2000.
[9] C. Wolf, J.M. Jolion and F. Chassaing, “Text localization, enhancement and binarization in multimedia documents,” Proc. of 16th Int. Conf. on Pattern Recognition, vol. 2, pp.1037-1040, Aug. 2002.
[10]H. Ancin and A.K. Bhattacharjya, “Text enhancement for laser copiers,” Proc. of Int. Conf. on Image Processing, vol. 3, pp.494-498, Oct. 1999.
[11]R.C. Gonzalez, R.E. Woods, Digital Image Processing, Second Edition, Prentice-Hall, Upper Saddle River, New Jersey, U.S.A., 2002.
[12]Y.C. Yeh, The Design and Application of Neural Network Model, 8th Edition, May 2000. (in Chinese)
[13]T. Naito, T. Tsukada, K. Yamada, K. Kozuka and S. Yamamoto, “License plate recognition method for inclined plates outdoors,” Proc. of Int. Conf. on Information Intelligence and Systems, pp.304-312, Oct. 1999.
[14]T. Sirithinaphong and K. Chamnongthai, “Extraction of car license plate using motor vehicle regulation and character pattern recognition,” The IEEE Asia-Pacific Conf. on Circuits and Systems, pp.559-562, Nov. 1998.
[15]M.J. Ahmed, M. Sarfraz, A. Zidouri, and W.G. Al-Khatib,“ License plate recognition system,” Proc. of IEEE Int. Conf. on Circuits and Systems, vol. 2, pp.898-901, Dec. 2003.
[16]F. Shen and H. Wang, "A local edge detector used for finding corners," Proc. of the Third Int. Conf. on Information, Communications and Signal, Oct. 2001.
[17]T. Naito, T. Tsukada, K. Yamada, K. Kozuka and S. Yamamoto, “Robust license-plate recognition method for passing vehicles under outside environment,” IEEE Trans. on Vehicular Technology, vol. 49, Issue 6, pp.2309-2319, Nov. 2000.
[18]Y.P. Huang, S.Y. La and W.P. Chuang, “A template-based model for license plate recognition,” IEEE Int. Conf. on Networking, Sensing and Control, vol. 2, pp.737-742, Mar. 2004.
[19]V. Koval, V. Turchenko, V. Kochan, A. Sachenko and G. Markowsky, “Smart license plate recognition system based on image processing using neural network,” IEEE Int. Workshop on Intelligent Data Acquisition and Advanced Computing Systems: Technology and Applications, pp.123-127, Sept. 2003.
[20]S.Z. Wang and H.J. Lee, “Detection and recognition of license plate characters with different appearances,” Proc. of IEEE Intelligent Transportation Systems, vol. 2, pp.979-984, Oct. 2003.
[21] D. Yan, H. Ma, J. Liu and L. Li, “A high performance license plate recognition system based on the web technique,” Proc. of IEEE Intelligent Transportation Systems, pp.325-329, Aug. 2001.
[22] S.L. Chang, L.S. Chen, Y.C. Chung and S.W. Chen, “Automatic license plate recognition,” IEEE Trans. on Intelligent Transportation Systems, vol. 5, Issue 1, pp.42-53, Mar. 2004.
[23] M. Yu and Y.D. Kim, “An approach to Korean license plate recognition based on vertical edge matching,” IEEE Int. Conf. on Systems, Man, and Cybernetics, vol. 4, pp.2975-2980, Oct. 2000.
[24] H.A. Hegt, R.J. de la Haye, and N.A. Khan, “A high performance license plate recognition system,” IEEE Int. Conf. on Systems, Man, and Cybernetics, vol. 5, pp.4357-4362, Oct. 1998.
[25] N.L.V. Sudhan, J. Ramkumar, and S.K. Priya, “New approaches for license plate recognition system,” Proc. of Int. Conf. on Intelligent Sensing and Information Processing, pp.149-152, Jan. 2004.