臺灣博碩士論文加值系統

現今的車牌偵測研究大部分都是針對近距離或是單一車道，或是針對前車來進行車牌的偵測與辨識，而在智慧型交通系統中道路監控是一個很重要的部分，但大部分研究要應用在道路監控上都不適用，因為如果對多車道畫面進行車牌偵測，會因為過多的干擾資訊導致偵測率下降，因此我們提出了能應用在多車道且單一車道也適用的快速車牌偵測系統，為了解決背景所造成的干擾，我們的架構分為兩層，先以局部二值圖樣(LBP)搭配適應性偵測範圍進行車輛的偵測來縮小我們要偵測的範圍，再根據所找到的車輛來，進行以邊緣偵測為基礎積分影像為輔助的遠距離車牌偵測。為了驗證我們系統的可行性，我們實際進行拍攝了10組台北市的道路畫面來進行測試，測試環境包含了晴天、陰天和雨天，以及單車道和多車道。車輛的偵測率平均是97%，而車牌的偵測率在單車道的影片中平均有94%，多車道則平均有93%的偵測率。

Most current license plate detection (LPD) researches target only on the vehicles on single-lane or in short distance to the camera, i.e. each frame contains only one vehicle, which are not suitable for Intelligent Transport Systems, because license plates detection directly on multi-lane would fail for the adopted edge features would be interfered by varying vehicles and background. Here, we proposed a LPD system which can be applied to both single-lane and multi-lane for road monitoring. In this paper, we proposed a system architecture which is based on local binary patterns (LBP) for vehicle detection and edge detection for license plate locating. By adaptive range detection and integral image, real time processing could be achieved for multi-lane license plate detection. In order to locate license plate, hierarchical approach which detect vehicle firstly is more efficiently. In order to verify the feasibility of proposed system, 10 video recoded in Taipei urban area including single-lane and multi-lane under sunny, cloudy, and rainy days are tested. The average vehicle detection rate is 97% and the LPD rate reached 94% and 93% for the single-lane and multi-lane, respectively.

目次 V
圖次 VII
表次 IX
第一章 導論 1
1.1. 研究動機與目標 1
1.2. 論文架構 3
第二章 相關研究 4
2.1. 近距離車牌偵測相關研究 4
2.2. 單車道車牌偵測相關研究 7
2.3. 車牌偵測相關研究總論 8
2.4. 車輛偵測相關研究 8
第三章 系統架構與實作 10
3.1. 架構設計與構思 10
3.2. 系統架構 11
3.3. LBP 12
3.4. 積分影像 13
3.5. CANNY EDGE DETECTOR 15
3.6. LBP的訓練模組 17
3.7. 車輛偵測模組 17
3.8. 車牌偵測模組 19
第四章 實驗成果 23
4.1. 車輛偵測結果 23
4.2. 車牌偵測結果 26
4.3. 近距與遠距測試 30
4.4. 錯誤類型分析與討論 33
第五章 結論與未來展望 35
參考文獻 36

[1]Ching-Tang Hsieh, Liang-Chun Chang, Kuo-Ming Hung and Hsieh-Chang Huang, “A real-time mobile vehicle license plate detection and recognition for vehicle monitoring and management,” Pervasive Computing (JCPC), 2009 Joint Conferences on, pp. 197-202, 2009
[2]Ran Wang, Guoyou Wang and Zhonghua Wang, “A Fast Method for Two-stage Precise License Plate Location Based on Fusion of Corner Feature and Gray Level Jumps,” Industrial Control and Electronics Engineering (ICICEE), 2012 International Conference on, pp. 976- 979, 2012
[3]Sheng Hao, Li Chao, Wen Qi and Xiong Zhang, “Real-Time Anti-Interference Location of Vehicle License Plates Using High-Definition Video,” Intelligent Transportation Systems Magazine, IEEE, Volume.1 , Issue.4, pp. 17- 23, 2009
[4]Hinde Anoual, Sanaa EL Fkihi, Abdellilah Jilbab and Driss Aboutajdine , “Vehicle License Plate Detection in Images,” Multimedia Computing and Systems (ICMCS), 2011 International Conference on, pp.1-5, 2011
[5]Yi-Chuan Liu and Chun-Hong Zheng, “Vehicle license plate location based on mathematical morphology and variance projection, ” Image Analysis and Signal Processing (IASP), 2011 International Conference on, pp. 360-363, 2011
[6]Kaushik Deb and Kang-Hyun Jo ,“HSI Color based Vehicle License Plate Detection,” Control, Automation and Systems,(ICCAS) 2008. International Conference on, pp. 687- 691, 2008
[7]Zuwena Musoromy, Faycal Bensaali, Soodamani Ramalingam and Georgios Pissanidis, “Comparison of Real-Time DSP-Based Edge Detection Techniques for License Plate Detection,” Information Assurance and Security (IAS), 2010 Sixth International Conference on, pp. 323- 328, 2010
[8]Chao-Ho Chen, Tsong-Yi Chen, Chi-Ming Huang, Da-Jinn Wang, “License Plate Location for Vehicles Passing through a Gate,” Intelligent Information Hiding and Multimedia Signal Processing (IIH-MSP), 2011 Seventh International Conference on, pp.340-343, 2011
[9]Xian Sheng and Shutao Li, “License plate detection based on improved Real Adaboost,” Vehicular Electronics and Safety (ICVES), 2011 IEEE International Conference on, pp. 169-173, 2011
[10]Wei Wang , Qiao-Jing Jiang , Xi Zhou and Wen-Yin Wan, “Car License Plate Detection Based on MSER,” Consumer Electronics, Communications and Networks (CECNet), 2011 International Conference on, pp. 3973-3976, 2011
[11]陳炳富, “智慧型交通監控系統中多車道車輛之偵測與追蹤,”
[12]高智源, “基於即時影像處理之交通資訊整合系統,”
[13]Wen-Chung Chang and Chih-Wei Cho, “Online Boosting for Vehicle Detection ,”IEEE TRANSACTIONS ON SYSTEMS, MAN, AND CYBERNETICS, Volume.40 , Issue.3, pp. 892 - 902, 2010
[14]Timo Ojala, Matti Pietikainen, David Harwood, "Performance evaluation of texture measures with classification based on Kullback discrimination of distributions," Proceedings of the 12th IAPR International Conference on Pattern Recognition (ICPR 1994), vol. 1, pp. 582 – 585 , 1996
[15]Timo Ojala, Matti Pietikainen, David Harwood, "A Comparative Study of Texture Measures with Classification Based on Feature Distributions," Pattern Recognition, vol. 29, pp. 51-5, 1996
[16]http://docs.opencv.org/modules/contrib/doc/facerec/facerec_tutorial.html