臺灣博碩士論文加值系統

本系統主旨為建置一個適合台灣環境使用之車牌辨識系統，必須要能濾除複雜背景、快速多標偵測汽車與機車車牌、校正車牌、車牌辨識。因此本系統分為四個部分:影像前處理、車牌偵測、車牌正規化、車牌辨識。
影像前處理部分本系統在能增加系統準確率與不增加處理時間之前提下利用車牌特有之色彩特徵提出色彩差異度測邊演算法結合Sobel橫向測邊而能夠濾除影像中約62.5%非車牌之邊點。車牌偵測部分尋找影像邊點中符合字元特性者再予以形態學處理找出車牌候選區，並在經由密度濾波與物件特性等特點找出實際車牌區域，此法可於單次掃描中找出影像中所有車牌，且不限於車牌配色，準確率高達96.41%，花費時間為168ms。車牌正規化部分為解決車牌具有歪斜或歪曲等不理想狀況，也為提升車牌辨識率所必須處理之過程，本文使用車牌字元斜度估計與歪曲車牌校正法，可校正之車牌為歪斜0°~36°，鑑於實際情況中歪斜超過30°之車牌極其罕見故為可接受之範圍。車牌辨識部分為增加字元辨識率，本系統使用投影法與切割區域估測法準確找出車牌中文字沾黏之部分再予以切割，且本系統使用機率模板文字比對法，比對字元每個像素值接有不同權重且執行速度快準確率也高達97.89%。最終本系統所達成之車牌辨識率為89.16%，平均系統花費時間為241ms。比較於現有其它已發表論文之偵測與辨識系統表現，本系統所受限制少且準確率較高。

This system is specially built for Taiwan’s environment. It has 4 parts: image preprocessing, multiple license-plate detection, normalization and recognition.
In image preprocessing, this system takes the benefit of license plate’s color feature then come out a color differential edge detection (CDED) algorithm combines with horizontal Sobel mask, which can erase 62.5% of uninteresting pixels. In detection part, finding edge pixel of character feature then using morphology and density filter extracts the real license plate region. This method can find all license plates region in one scan and has 96.41% succession rate therefore consuming time is 168ms. Normalization part has to solve license plate’s rotation and warp to increase recognition succession rate. This system uses characters in license-plate to estimation rotation angle, and can normalize angle of 0°~36°. In recognition part, projection segmentation combines with segmentation region estimation method can segment sticky characters and increase recognition rate. Finally this system uses probability templates as OCR which has weight for each pixel for every character, and it is fast and has 97.89% succession rate. Of all, this System’s recognition rate is 89.16% and consuming time is 241ms. This paper also collects nowadays LPR related papers, and this system has less restricts and higher succession rate than those.

論文摘要 I
Abstract II
致 謝 III
目 錄 IV
圖目錄 VIII
表目錄 XI
第一章 緒論 1
1-1 研究動機 1
1-2 研究目標 1
1-3 研究方法 2
1-4 論文架構 3
第二章 文獻探討 5
2-1 數位影像處理基礎 5
2-1.1 影像色彩模型 5
(A) RGB 5
(B) HSV 7
2-1.2 灰階 8
2-1.3 二值化 8
(A) 固定式門檻值 8
(B) 統計式門檻值－Otsu 8
2-1.1 影像濾波器 10
(A) 平滑濾波器 11
(B) 銳化濾波器 14
2-1.2 邊緣檢測 15
(A) Sobel 16
(B) Canny 17
2-1.3 形態學 18
(A) 膨脹 19
(B) 侵蝕 20
(C) 斷開 21
(D) 閉合 21
2-1.4 輪廓追蹤之物件擷取 22
2-1.5 影像歪曲校正 23
2-2 車牌辨識技術探討 24
2-2.1 車牌偵測 25
(A) 投影量分析 26
(B) 高密度邊點區域搜尋 27
(C) 色彩特徵 30
(D) 形態學與連通物件分析 31
2-2.2 車牌正規化 33
2-2.3 車牌辨識 34
(A) 字元切割 34
(B) 文字辨識(OCR) 35
2-3 目前車牌辨識方法比較 38
第三章 本系統架構 42
3-1 系統流程 42
3-2 影像前置處理 44
3-2.1 色彩差異度邊緣檢測 45
3-3 車牌偵測 48
3-3.1 文字特徵 48
(A) 文字邊點物件追蹤 48
(B) 水平物件連結 49
3-3.2 車牌特徵 49
3-4 車牌正規化 50
3-4.1 車牌大小校正與二值化 51
3-4.2 車牌字元擷取 54
3-4.3 車牌旋度校正 55
3-4.4 邊界切割 56
3-4.5 四角點扭曲校正 57
3-5 車牌辨識 58
3-5.1 字元切割 59
3-5.1 字元辨識 60
3-5.2 易混淆字型判定 62
(A) 0 &; D 62
(B) 8 &; B &; R 63
(C) 1&;I 63
第四章 研究結果 65
4-1 系統環境架構 65
4-2 車牌偵測與辨識系統測試數據 65
4-3 執行速度 68
4-4 實驗結果範例 68
4-4.1 拍攝角度 70
4-4.2 不同拍攝環境與車牌種類 71
4-5 車牌偵測與辨識失敗討論 75
4-5.1 偵測失敗 75
4-5.2 辨識失敗 77
第五章 結論與未來展望 78
5-1 結論 78
5-2 未來展望 79
參考文獻 80

[1] N. Otsu, “A Threshold Selection Method from Gray-level Histograms,” IEEE Systems Man and Cybernetics (1979).
[2]鍾國亮,“影像處理與電腦視覺,”第四版, 台灣東華書局(2008)。
[3]莊佳龍,“車輛偵測與車牌辨識系統–Vehicle Detection and License Plate Recognition System,”碩士論文,中正大學(2005)。
[4] W. Burger, M.J. Burge, “Digital Image Processing－An algorithmic Introduction using Java,”1st ed , Springer, (2007).
[5] M.Y. Dehkordi, M. Nikzad, V.R. Ekhlas, Z. Azimifar, “ A Novel Approach for Fast and Robust Multiple License Plate Detection,” Machine Vision and Image Processing (2010).
[6] C.C. Lin , W.H. Huang, “Locating License Plate Based on Edge Features of Intensity and Saturation Subimages,” Innovative Computing Information and Control, Kaohsiung (2007).
[7]張瑋珊,“動態視訊之車牌辨識系統,” 碩士論文, 台灣科技大學, (2011)。
[8] C.N. Paunwala, “Multiple License Plate Extraction Based on Mathematical Morphology and Component Filtering in Indian Traffic Conditions,” Advances in Recent Technologies in Communication and Computing (2010).
[9]C. Wu, L.C. On, C.H. Weng, T.S. Kuan, and K. Ng, “A Macao license plate recognition system,” Machine Learning and Cybernetics, China, (2005).
[10]R. C. Gonzalez and R. E. Woods, “Digital Image Processing,” 3rd ed, Addison-Wesley, MA (1993).
[11]B. Hongliang, L. Changping, “A Hybrid License Plate Extraction Method Based on Edge Statistics and Morphology,” Pattern Recognition, 2004. ICPR 2004. Proceedings of the 17th International Conference , vol.2, pp831-834 (2004).
[12]N. Bellas, S.M. Chai, M. Dwyer, and D. Linzmeier, “FPGA Implementation of a License Plate Recognition SoC Using Automatically Generated Streaming Accelerators,” 20th IPDPS, France (2006).
[13] C. Oz and F. Ercal, “A Practical License Plate Recognition System for Real-time Environments,” IWANN, vol. 3512. Berlin, Germany: Springer-Verlag (2005).
[14]L. Angeline, “Multiple Vehicles' License Plate Tracking and Recognition via Isotropic Dilation,” Computational Intelligence, Communication Systems and Networks (CICSyN) (2011).
[15] I.E Anagnostopoulos and I.D. Psoroulas , V. Loumos, E. Kayafas , “License Plate Recognition from Still Images and Video Sequences: A Survey,” IEEE Intelligent Transportation Systems. (2008).
[16] J.M. Lopez, “A Versatile Low-cost Car Plate Recognition System,” Signal Processing and Its Applications (2007).
[17] C.N.E. Anagnostopoulos, I.E Anagnostopoulos, V. Loumos and E. Kayafas ,” A License Plate-Recognition Algorithm for Intelligent Transportation System Applications,” IEEE Intelligent Transportation Systems, vol.7, pp377-392.(2006).
[18]簡維皇, ”用於路口查驗贓車的車牌辨識系統A License Plate Recognition System for Finding Out the Stolen Vehicles,”碩士論文,中央大學(2011)。
[19]S.-L. Chang, L.-S. Chen, Y.-C. Chung, and S.-W. Chen, “Automatic license plate recognition,” IEEE Trans. Intell. Transp. Syst., vol. 5, no. 1, pp. 42–53 (2004).
[20]A. Broumandnia and M. Fathy, “Application of Pattern Recognition for Farsi License Plate Recognition,” ICGST Int. Conf. Graphics, Vision and Image Processing (2005).
[21]F. Kahraman, B. Kurt, and M. Gokmen, “License Plate Character Segmentation Based on the Gabor Transform and Vector Quantization,” Lecture Notes on Computer Science, vol. 2869, pp. 381–388 (2003).
[22] T. D. Duan, T. L. Hong Du, T. V. Phuoc, and N. V. Hoang, “Building an Automatic Vehicle License Plate Recognition System,” Conf. Comput. Sci. RIVF, pp. 59–63 (2005).
[23]K. K. Kim, K. I. Kim, J. B. Kim, and H. J. Kim, “Learning-based Approach, for License Plate Recognition,” IEEE Signal Process. Soc. Workshop, Neural Netw. Signal Process , vol. 2, pp. 614–623 (2000).
[24] H.-J. Lee, S.-Y. Chen, and S.-Z. Wang, “Extraction and Recognition of License Plates of Motorcycles and Vehicles on Highways,” Proc ICPR, pp. 356–359 (2004).
[25]J. Cano and J. C. Perez-Cortes, “Vehicle License Plate Segmentation in Natural Images,” Lecture Notes on Computer Science, vol. 2652, pp. 142–149 (2003).
[26]X. Shi,W. Zhao, and Y. Shen, “Automatic License Plate Recognition System Based on Color Image Processing,” in Lecture Notes on Computer Science, vol. 3483, pp. 1159–1168 (2005).
[27] M.K. Hu, “Visual Pattern Recognition by Moment Invariants,” IRE Trans. Info. Theory, vol. IT-8, pp179-187 (1962).
[28]陳祥傑,” 自動化車牌辨識系統設計,” 碩士論文, 中央大學(2005)。.
[29]賴琮凱”多標的車牌辨識系統之研製,” 碩士論文, 建國科技大學(2008)。
[30]魏銪志”動態多標的車牌辨識系統之研究.” 碩士論文, 元智大學(2000)。
[31]王振興”多標的汽機車車牌辨識系統之研究,” 碩士論文, 元智大學(2003)。
[32]陳光中,”及時車牌辨識系統-動態影像中有效辨識多個車牌的方法,” 碩士論文,台北科技大學(2010)。
[33]陳家豪,”基於Kinect 之境/動態機車車牌辨識,” 碩士論文, 淡江大學(2012)。
[34]梁智凱,”多車牌辨識系統之研究,” 碩士論文, 淡江大學(2006)。
[35]劉國偉,”移動中車牌定位系統之設計與實作,” 碩士論文, 台北科技大學(2005)。
[36]吳建忠,”於未設限環境中車牌辨識系統模擬與分析,” 碩士論文, 台灣科技大學( 2006)。
[37]蔡明翰,”超速車輛之車牌定位及字元分割,” 碩士論文, 台灣大學(2009)。
[38]蘇仲洋,”影像辨識在全自動化系統之研究,” 碩士論文, 台北科技大學(2008)。
[39]莊祥琳,”動態車牌辨識系統之研究,” 碩士論文, 淡江大學(2007)。
[40]黃漢強,”具備多重辨識器並以PDA為平台之車牌辨識系統,” 碩士論文, 中原大學(2007)。