本論文以實現嵌入式平台為設計考量，選用運算複雜度較低的影像處理技術或演算法，盡可能利用簡單的數學運算減少運算時間，以降低嵌入式系統負擔。影像處理程序分成定位處理階段、前置處理階段及辨識處理階段。定位處理階段的工作就是擷取車牌的部份，其處理步驟為灰階轉換、影像強化、邊緣偵測、密度計算及車牌定位；前置處理階段的目的是輔助提升字元辨識率，其處理步驟為影像二值化、車牌四邊頂點估測、車牌傾斜矯正、影像細化及字元分割；辨識處理階段採用倒傳遞神經網路做為字元辨識演算法，其處理步驟為字元正規化及字元辨識。本論文採用1026張大小為384x256接近現實未設限環境影像進行驗證，將影像依其車牌亮度、影像複雜度、車牌拍攝距離、車牌拍攝角度、車牌傾斜角度及車牌周圍色調等特性加以分類。依照其正確率探討失敗步驟與原因、字元辨識率及平均執行時間，評估系統效能並且提供影像處理步驟或演算法修改前後比較的基準。利用分析結果找出系統在設限環境的屬性、探討未設限環境對系統造成的影響及針對可能面臨的問題找出可行的解決方案。最後找出適當的車牌定位與字元辨識系統之影像處理程序，以期在未設限環境中增加系統應用領域或實用性，使其更符合實際應用。

In this thesis, we design a method for the localization and recognition of vehicle license plates system. In order to be implemented on an embedded platform, we apply algorithm with simple mathematical operation to reduce the require computation. The proposed method consist of three parts: localization, preprocessing, and recognition. Localization seeks out probable position of vehicle license plates. Preprocessing helps next stage to increase recognition rate. It applies back propagation neural networks to identify each character of vehicle license plates in recognition. We take 1026 images in unrestricted environment each with 384x256 pixel to test our method. These images are classified by illumination, complexity, distance, shooting angle, tilted angle of vehicle license plates, and color of vehicle license plates background. Then, we analyze the recognition rate, average execution time, and reasons for failure in different image classes. Also, the efficiency of each step in our algorithm is studied. It provides us with information about the critical steps. This will be useful to improve the system in the future.

[1]. 黃振欣，“應用倒傳遞網路模型設計PDA車牌辨識系統”，大同大學，資訊工程研究所碩士論文，民國九十三年六月。
[2]. 陳麗奾，“在未設限的環境下車牌的定位與辨識”，國立師範大學，資訊教育研究所碩士論文，民國八十九年六月。
[3]. 李國彥，“車牌辨識系統之研究”，大同大學，通訊工程研究所碩士論文，民國九十年六月。
[4]. 官宗保，“利用數位訊號處理器實現車牌字元辨識系統”，國立台灣大學，電機工程學研究所碩士論文，民國九十年六月。
[5]. 陳朕寬，“彩色車牌的自動偵測方法”，國立中山大學，機械與機電工程學研究所碩士論文，民國九十四年七月。
[6]. H. J. Lee, D. H. Kim, D. J. Kim, and S. Y. Bang, “Real-time automatic vehicle management system using vehicle tracking and car plate number identification,” International Conference on Multimedia and Expo, vol.2, 2003, pp.353-356.
[7]. Q. C. Tao, X. H. He, D. S. Luo, and W. Wu, “A new car plate recognition method based on fuzzy entropy,” Fifth World Congress on Intelligent Control and Automation, vol.5, 2004, pp.4054-4056.
[8]. H. L. Bai, J. M. Zhu, and C. P. Liu, “A fast license plate extraction method on complex background,” IEEE on Intelligent Transportation Systems, vol.2, 2003, pp.985-987.
[9]. C. Coetzee, C. Botha,and D. Weber, “PC based number plate recognition system,” IEEE International Symposium on Industrial Electronics, vol.2, 1998, pp.605-610.
[10]. J. A. G. Nijhuis, M. H. Ter Brugge, K. A. Helmholt, J. P. W. Pluim, L. Spaanenburg, R. S. Venema and M. A. Westenberg, “Car license plate recognition with neural networks and fuzzy logic,” IEEE International Conference on Neural Networks, vol.5, 1995, pp.2232-2236.
[11]. M. H. Ter Brugge, J. H. Stevens, J. A. G. Nijhuis and L. Spaanenburg, “License plate recognition using DTCNNs,” IEEE International Workshop on Cellular Neural Networks, 1998, pp.212-217.
[12]. 葉怡成，“類神經網路模式應用與實作”，台北，儒林書局，民國八十二年。
[13]. 張斐章，張麗秋，黃浩倫，“類神經網路理論與實務”，台北，東華書局，民國九十二年。
[14]. A.S. Pandya and R.B. Macy, Pattern Recognition with Neural Networks in C++, Florida, CRC PRESS, 1996.
[15]. J.S.R. Jang, C.T. Sun and E. Mizutani, Neuro-Fuzzy AND Soft Computing, New Jersey, Prentice-Hall, 1997.
[16]. 林丕靜，“數值分析”，台北，儒林書局，民國七十九年。
[17]. 陳廷標，夏良正，“數位影像處理”，台北，儒林書局，民國八十二年。
[18]. 鍾國亮，“影像處理與電腦視覺”，台北，東華書局，民國九十三年。
[19]. L.G. Shapiro and G.C. Stockman, Computer Vision, New Jersey, Prentice-Hall, 2001.
[20]. R.C. Gonzalez and R.E. Woods, Digital Image Processing, Massachusetts, Addison-Wesley, 1993.
[21]. J. Duan and G. Qiu, “Novel histogram processing for colour image enhancement,” International Conference on Image and Graphics, 2004, pp.55-58.
[22]. B. Enyedi, L. Konyha, C. Szombathy and K. Fazekas, “Strategies for fast license plate number localization,” International Symposium Electronics in Marine, 2004, pp.579-584.
[23]. D. Trier and A. K. Jain, “Goal-directed evaluation of binarization methods,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol.17, 1995, pp.1191-1201.
[24]. T. Y. Zhang and C. Y. Suen, “A fast parallel algorithm for thinning digital patterns,” Communications of the ACM, vol. 27, 1984, pp.236-239.
[25]. J. S. Kwon, J. W. Gi and E. K. Kang, “An enhanced thinning algorithm using parallel processing,” International Conference on Image Processing, vol.3, 2001, pp.752-755.