跳到主要內容

臺灣博碩士論文加值系統

(44.192.67.10) 您好!臺灣時間:2024/11/09 18:34
字體大小: 字級放大   字級縮小   預設字形  
回查詢結果 :::

詳目顯示

我願授權國圖
: 
twitterline
研究生:陳錫卿
研究生(外文):Hsi-Ching Chen
論文名稱:於高速道路下之車牌定位與辨識之研究
論文名稱(外文):The Research of Vehicle License Plates Moving Object Position and Recognition for The Highway
指導教授:周碩彥周碩彥引用關係
指導教授(外文):Shuo-Yan Chou
學位類別:碩士
校院名稱:國立臺灣科技大學
系所名稱:自動化及控制研究所
學門:工程學門
學類:機械工程學類
論文種類:學術論文
論文出版年:2003
畢業學年度:91
語文別:英文
論文頁數:69
中文關鍵詞:智慧型交通運輸系統光學辨識交錯性車牌定位車牌辨識
外文關鍵詞:ITSOCRInterlacePositionRecognition
相關次數:
  • 被引用被引用:3
  • 點閱點閱:1184
  • 評分評分:
  • 下載下載:0
  • 收藏至我的研究室書目清單書目收藏:1
自動車牌辨識系統可以廣泛的運用在一些自動化系統理,例如:收費停車場管理、高速公路收費站、贓車的查緝以及自動收費索償等電子收費系統等。一個自動收費系統有著高且可靠的辨識率,不但可以做到無人控管,還可以節省人工管理費用,ITS (Intelligent Transportation system)已成為未來的趨勢,我們希望在未來能透過此方法輔助整個ITS。
近幾年來,國內外數位影像處理在文數字辨識能力已經有明顯的增加。所以,一個完整的車牌辨識系統是整合三個主要的步驟結合而成的,包括影像前處理、車牌定位和車牌辨識。此外,在影像復原和辨識演算法依然有改進的空間。因此,本論文引用了一個新的技術,像在統計的BLUE (best linear unbiased Estimator)和代數裡的矩陣方式去改善動態影像所造成的模糊問題。
本文運用了三種二質化的方法,這些方法將解決任何影像的問題。有了這些影像前處理得過程然後使用隨機法找出四條直線去定位。不只如此,本論文創造了辨識演算法去辨識車牌字元。
最後,我們的辨識系統一共實驗了948張影像,其中包括了492張靜態的圖片,455張動態的圖片。本系統使用Borland C++ Builder 5.0,CPU K6II 400 MHz和128M的RAM。整體的辨識率達到92.3%。我們的取像設備採用Kodak DX3900 Zoom 兩倍光學和SONY DCR-PC3 NTSC Mini DV 40X Carl Zeiss。
An automatic vehicle recognition system that can be widely used in the system of automation, such as management of parking lot, lots of freeways, tracking stolen vehicles, automatic fee charge…etc. An automatic system with high recognition accuracy and reliability, it will achieve the goal of wising management and save the wage of guards, such as ITS (Intelligent Transportation System), it will become the trend in the feature.
Recently, the rapid development of digital image processing technique has increased the ability of alphanumeric recognition. Certainly, a complete vehicle license plate recognition system is integrated by three main including preprocessing of image, vehicle license plate position, recognition of vehicle license plate. Moreover, there still exist many improving area at image retrieve technique and recognition algorithm. Therefore, in this thesis, we propose some new techniques such as BLUE in statistics and matrix in algebra to improve the misty problem that dynamic state image caused when the vehicle of picking and speed up we want to recognize.
In this thesis we have three methods of bi-level, this method we will solve any situation of images. Having this image preprocess, then using random method to find four straights for position. This thesis creates algorithm of recognition to recognize the characters in the vehicle license plate.
. Finally, Our vehicle plate recognition system is experimented with total 948 images among 492 for static state vehicle, 456 for dynamic state vehicle This system algorithm is programming by Borland C++ Builder 5.0, use K6II 400MHz and 128MB RAM. The whole recognition ratio is 92.3%, We use Kodak DX3900 Zoom which is optical 2X digital camera and SONY DCR-PC3 NTSC Mini DV 40X Carl Zeiss 20 ten-thousand pixels.
ABSTRACT I
ACKNOWLEDGEMENTS III
CONTENTS IV
LIST OF FIGURES VII
LIST OF TABLES VIII
CHAPTER 1 INTRODUCTION 1
1.1 BACKGROUND 1
1.2 MOTIVE 2
1.3 OBJECT AND PROBLEM ILLUSTRATES 5
1.3.1 Object 5
1.3.2 The problem illustrates 5
1.3.2.1 The dynamic state cause the faintness 6
1.3.2.2 Similar to the image of the vehicle license plate 6
1.3.2.3 The license plate contamination, frame changes and decorates 6
1.3.2.4 The image’s angle too large 6
1.3.2.5 The strong light or images contrast lacked 7
1.3.2.6 The image of vehicle plate too small 7
1.4 ASSUMPTIONS AND RESTRICTIONS 8
1.4.1 Assumptions 8
1.4.2 Restrictions 8
1.5 METHODS OF RESEARCH 8
CHAPTER 2 PREPROCESSING 12
2.1 STRUCTURAL THE FLOW CHART 12
2.2 LITERATURE REVIEW 13
2.2.1 The result of researching to Sin-Ping Lin 13
2.2.2 The result of researching to Chung-Shih Jong 13
2.2.3 The result of researching to Wen-Zong Yang 14
2.2.4 The result of researching to Jin-Yi Huang 14
2.2.5 The result of researching to You-Zhi Wei 15
2.2.6 The result of researching to Jhy-Hong Juang 15
2.3 PRE-PROCESSING 15
2.3.1 The description of problems 16
2.3.2 Gray Level Transformation of image 16
2.3.3 Filters 17
2.3.3.1 Preface 17
2.3.3.2 Mean Filter 17
2.3.3.3 Median Filter 19
2.3.3.4 Homomorphic Filter 19
2.4 BI-LEVEL 22
2.4.1 The method for threshold value 22
2.4.2 Bi- level basic concept and result 23
2.4.3 Three kinds of methods 23
2.4.3.1 Otsu’s Bi-level method 23
2.4.3.2 User-define Method 25
2.4.3.3 Average value Bi-level 26
2.4.3.3.1 Histogram Modification 26
2.4.3.3.2 Histogram equal parts 26
CHAPTER 3 MOVING OBJECT PROCESSING 30
3.1 PREFACE 30
3.2 DE-INTERLACE PHENOMENON 30
3.3 IMAGE RESTORATION 33
3.3.1 Statement 33
3.3.2 Methods 33
3.3.3 The image reduces the quality system 34
3.3.3.1 Basic definition 34
3.3.4 The method of algebra restores 35
3.4 LEAST SQUARE METHOD FOR IMAGE RESTORE 36
3.5 EXPERIMENT RESULTS: 38
CHAPTER 4 EDGE DETECTION AND POSITION 41
4.1 EDGE DETECTION 41
4.1.1 Relative background knowledge 41
4.1.2 The operator category of edge detection 41
4.1.2.1 Gradient operand 42
4.1.2.2 Laplacian Operator 44
4.2 POSITION 48
4.2.1 Preface 48
4.2.2 The method of search the edge straight 48
4.2.2.1 The method of force random choice 48
4.2.2.2 Algorithm of position 50
4.3 THE RESULT OF POSITION 52
CHAPTER 5 THE RECOGNITION OF THE VEHICLE LICENSE PLATES 54
5.1 INTRODUCTION 54
5.2 RECOGNIZE THE PRE-PROCESSING 54
5.2.1 Preface 54
5.2.2 Geometry correction of the image flat 55
5.3 CATCHING THE CHARACTER OF THE LICENSE PLATE 56
5.4 THE RECOGNITION OF THE VEHICLE LICENSE PLATE 58
5.4.1 Algorithm of recognition 59
5.4.1.1 Effectively for recognition 62
5.4.1.2 Represents for Binary-Span-Tree 62
5.4.1.3 Diagram for Binary-Span-Tree 63
5.5 THE RESULT FOR RECOGNITION 64
CHAPTER 6 CONCLUSION 67
6.1 THE RESULT OF EXPERIMENTATION 67
6.2 SYSTEM EVALUATE 67
6.3 CONTRIBUTION 68
6.4 FUTURE WORKS AND CONCLUSION 68
6.4.1 Future works 68
6.4.2 Conclusion 69
BIBLIOGRAPHY R
APPENDIX A A
[1] Jhy-Hong Juang, The Recognition of Vehicle Plate Characters Using WMMS Method, National Tsing Hua University (NTHU),Electrical Engineering, in Chinese, 1999.
[2] Jong-Jhy Tash, Character Partition. Normalization and Recognition Research of License Plate Recognition Syste, National Tsing Hua University (NTHU), Electrical Engineering, in Chinese, 1999.
[3] Yang-Jian Zeng, A Research on Plate Localization Preprocessing and Recognition of License Plate Recognition System, National Tsing Hua University (NTHU), in Chinese, Electrical Engineering,1999.
[4] Wan-Chi Zheng, Application of Neural Networks in Handwritten Digit Recognition, Tamkang University (TKU), Electrical Engineering, in Chinese, 2000.
[5] Jin-Yi Huang, Design of the Vehicle License Plate Recognition System, Tamkang University (TKU), Electrical Engineering, in Chinese, 2000.
[6] Li-Xian Chen, Localization and Recognition of Vehicle License Plates in Unrestricted Environments, National Taiwan Normal University (NTNU), Information & Computer Education, in Chinese, 2000.
[7] Sin-Ping Lin, The Extraction of Vehicle Plate Characters, National Chiao Tung University (NCTU), Electrical and Control Engineering, in Chinese, 1999.
[8] Chung-Shih Jong, A Hybrid Approach for Automatic Vehicle License Plate Recognition, Yuan Ze University (YZU), Computer & Science Engineering, in Chinese, 1998.
[9] Xi-Cong Ma, Vehicle Plate Recognition Using Gray Relational Grade, National Taiwan University of Science and Technology(NTUST), Electrical Engineering, in Chinese, 1997.
[10] Jia-Wen Li, Vehicle Plate Recognition System, National Taiwan University of Science and Technology (NTUST), Electrical Engineering, in Chinese, 2000.
[11] Yi-Kai Chen, Preprocessing of Document Images for Automatic Character Recognition, National Cheng Kung University (NCKU), Computer Science and Information Engineering, in Chinese, 2000.
[12] Shi-Zhang Liang, Use of Genetic Algorithms in Trainning of Fuzzy Neural Network, National Chiao Tung University (NCTU), Electrical and Control Engineering, in Chinese, in Chinese, 1999.
[13] He-Ren Lin, Microsoft Windows Implementation Of A Mandarin Phrase Recognition System, National Sun Yat-sen University (NSYSU), Electrical Engineering, in Chinese,1999.
[14] Ke-Cong Li, Neural Network combined with Genetic Algruthm-Evolutionary Neural Network, National Chiao Tung University (NCTU), Electrical and Control Engineering, in Chinese, 1999.
[15] Guan Zong, DSP Embedded System for Automatic Licence Plate Recognition, National Taiwan University (NTU), Electrical Engineering, in Chinese, 2001.
[16] You-Zhi Wei, Montion Based Multiple Vehicle License Plate Recognition, Yuan Ze University (YZU), Computer & Science Engineering, in Chinese.
[17] Wen-Zong Yang, Automatic Vehicle Identification Number Recognition System, Chung Yuan Christian University (CYCU), Electronic Engineering, in Chinese.
[18] Jinnun-Nan Chow, License Plate Recognition System, National Sun Yat-sen University (NSYSU), Electrical Engineering, in Chinese, 1995。
[19] Yan-Zheng Xu,Xin-Jin Yei,Jian-Ming Chen, Using Motorola and Intel CPU to evaluation and comparison in the license plate recognition system, http://www.iot.gov.tw/, 1996。
[20] Zhong-Mu Huang, Shi-Yong Xu, http://twp.apipa.org.tw/, TWP, Serial numbers:191905。
[21] Zhong-Mu Huang, Shi-Yong Xu, Wen-Ming Wu, http://twp.apipa.org.tw/, TWP, Serial numbers:221193。
[22] Zhong-Mu Huang, Shi-Yong Xu, http://twp.apipa.org.tw/, TWP, Serial numbers:221193。
[23] Tian-Gui Li,” Automatic Vehicle License Plate Recognition System”, Mechanical Industry Magazine,1990,pp196-205.
[24] M.Sonka,V.Hilavac,R.Boyle,Image Processing,Analysis,and Machine Vision,PWS Publishing,1998.
[25] Ching-Tang Hsieh,You-Chuang Wang and Ching-I Huang,”An Effective Method for Optical Recongnition of Vehicle License Plates”
[26] M.A.Fischler and R.C.Bolles,”Random sample consensus:A paradigm for model fitting with applications to image analysis and automated cartography,”Commun,ACM,24(6),1981,pp.381-395
[27] L.Xu,E.Oja and.Kultanan,”A new curve detection method:randomized Hough transform (RHT),”pattern Recongnittion Letters,11(5),1990,pp.331-338
[28] T.C.Chen and K.L.Chung,”An new randomized algorithm for detecting lines,”Real-Time Imaging,7(6),2000,pp.473-482.
[29] Run-Haw Yang,”License Plate Numberal Identification In A Car Image”,National Chiao Tung University,ROC,1987.
[30] Chi-Wai Chang.Wen-Thong Chang,”An Expert System for Chinese Newspaper Segmentation”,1991 Workshop.
[31] Yuan-Hung Leu and Ming-Yang Chern,”Detection of Zebre Crossing for Autonomous Vehicle Navigation”,1991 Workshop on CVGIP.
[32] Rafael C. Gonzalez Richared E. Woods, “Digital Image Processing”, Addision Wesley Pushing Company,1992.
[33] N. Otsu, “A Threshold selection method from grey-level histograms”, IEEE Trans. Syst., Man Cybern., vol. SMC-8, pp62-66,1978.
[34] J. R. Cowell, “Syntactic pattern recognizer for vehicle identification numbers”, Image and Vision Computing, Vol. 13, no, 1, pp.13-19,1995.
[35] P. Comelli, P. Ferragina, M. N. Granieri and F. Stabile, “Optical recognition of motor Vehicle licence plates,” IEEE Trans.Vehicular Technology, Vol.44, no.4, pp.790-799, November 1995.
[36] Motor Vehicle licence plates,” IEEE Trans.Vehicular Technology, Vol.44, no.4, pp.790-799, November 1995.
[37] Comelli, Paolo, Paolo Ferragaina, Mario Notturno Granieri and Flavio Stabile, “Optical Recognition of Motor Vehicle License Plates,” IEEE Transaction on Vehicular Technology, Vol.44, No.4, pp. 790-799
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top