臺灣博碩士論文加值系統

駕駛輔助系統的目的是希望利用不同的自動化技術來協助提升駕駛人之行車安全。路標辨識系統在整個駕駛輔助系統中一直都扮演著重要角色之一，因為它可以提供駕駛人獲得車輛前方的路標資訊。本文提出了一個結合兩種不同分類器之路標辨識系統，用以對台灣道路上常見的交通標誌進行偵測及辨識，整個系統包含路標偵測及路標辨識兩個主軸。在路標偵測方面，它使用路標之色彩資訊結合Adaboost 分類器的方式來進行路標偵測的工作，在路標內容辨識方面，則先利用Canny 邊緣偵測法形成特徵向量，再採用支援向量機(support vector machine, SVM)方法來進行路標內容之辨識。本系統也針對諸如亮度偏暗、視角偏轉、部分遮蔽和存在太多紅色資訊等干擾取得正確偵測之問題提供解決之技巧。經由實驗結果顯示，本文所設計之系統即使是對處理動態影片也能在可接受之辨識率下，表現出很好的執行效率。

Many different automatic technologies have been used to develop driver assistance systems for improving the safety of driving. Road sign recognition system is an important subsystem of a driver assistance system. It can be used to provide the driver about the road sign information in front of the vehicle. In this thesis, a road sign recognition system was proposed which combined Adaboost classifier and support vector machine(SVM) to do the road sign detection and the content recognition, respectively. In the content representation phase, the Canny edge detection method was adopted to obtain the feature vector, to be recognized by SVM, of the detected road sign image. The proposed system can detect the road signs correctly for the captured image under the conditions such as low illumination, rotation, occlusion and rich red color. From the experimental results, it is shown that the proposed system can perform well, under accepted recognition rate, for video input. It is encouraged to apply the proposed system to a real time application.

摘要I
Abstract II
誌謝III
目錄IV
圖目錄VI
表目錄VII
第一章 緒論1
1.1 研究動機與目的1
1.2 相關文獻回顧5
第二章 分類器相關研究探討17
2.1 Adaboost 分類器17
2.2 支援向量機(Support Vector Machine)24
2.2.1 線性可分離之資料分佈28
2.2.2 線性不可分離之資料分佈32
第三章 路標偵測與辨識38
3.1 路標偵測階段42
3.1.1 前置處理44
3.1.2 訓練Adaboost分類器50
3.2 內容辨識階段53
3.2.1 產生特徵向量55
3.2.2 訓練支援向量機58
第四章 實驗結果與討論62
4.1 靜態影像測試62
4.2 動態影片測試67
第五章 結論與未來展望72
5.1 結論72
參考文獻75

[1] 交通部國家交通統計比較http://www.motc.gov.tw/mocwebGIP/wSite/ct?xItem=4881&ctNode=169&mp=1, June 2009.
[2] 內政部警政署統計資料http://www.npa.gov.tw/NPAGip/wSite/np?ctNode=11358&mp=1, June 2009.
[3] 高速公路局智慧化運輸系統專區http://www.freeway.gov.tw/Publish.aspx?cnid=590, June 2009.
[4] 楊修銘, "干擾狀況下的交通標誌偵測與辨識" 國立政治大學資訊科學學系碩士論文, 2004.
[5] 鄭世聰, "鄉野道路之路邊界偵測及交通標誌辨識" 國立中正大學電機工程研究所碩士論文, 2003.
[6] 薛書倫, "兩階段式交通標誌辨識" 元智大學資訊工程學系碩士論文, 2002.
[7] 謝宗丞, "路標辨識系統" 義守大學工業工程與管理學系碩士論文, 2005.
[8] Y.-S. Liu, D.-J. Duh, S.-Y. Chen, R.-S. Liu, and J.-W. Hsieh, "Scale and skew-invariant road sign recognition," International Journal of Imaging Systems and Technology, vol. 17, pp. 28-39, 2007.
[9] S. Maldonado-Basc�曝, S. Lafuente-Arroyo, P. Gil-Jim�聲ez, H. G�曠ez-Moreno, and F. L�櫝ez-Ferreras, "Road-Sign Detection and Recognition Based on Support Vector Machines," IEEE Transactions on Intelligent Transportation Systems, vol. 8, pp.264-278, 2007.
[10] Y.-Y. Nguwi and A. Kouzani, "Detection and classification of road signs in natural environments," Neural Computing & Applications, vol. 17, pp. 265-289, 2007.
[11] J. Miura., T. Kanda., and Y. Shirai., "An active vision system for real-time traffic sign recognition," in Intelligent Transportation Systems. Proceedings. IEEE, pp.52-57, 2000.
[12] X. W. Gao, L. Podladchikova, D. Shaposhnikov, K. Hong, and N. Shevtsova, "Recognition of traffic signs based on their colour and shape features extracted using human vision models," Journal of Visual Communication and Image Representation, vol. 17, pp. 675-685, 2006.
[13] N. Barnes, A. Zelinsky, and L. S. Fletcher, "Real-Time Speed Sign Detection Using the Radial Symmetry Detector," Intelligent Transportation Systems, IEEE Transactions on, vol. 9, pp. 322-332, 2008.
[14] C. Yoon, H. Lee, E. Kim, and M. Park, "Real-Time Road Sign Detection Using Fuzzy-Boosting," IEICE Trans Fundamentals, vol. E91-A, pp. 3346-3355, 2008.
[15] A. Koncar, H. JanBen, and S. Halgamuge, "Gabor wavelet similarity maps for optimising hierarchical road sign classifiers," Pattern Recognition Letters, vol. 28, pp.260-267, 2007.
[16] M. Shi, H. Wu, and H. Fleyeh, "Support vector machines for traffic signs recognition,"in Neural Networks, IJCNN. (IEEE World Congress on Computational Intelligence),pp. 3820-3827, 2008.
[17] P. Viola and M. Jones, "Robust Real-time Object Detection," ICCV 2nd International Workshop on Statistical and Computational Theories of Vision : Modeling, Learning, Computing and Sampling, July 2001.
[18] Y. Freund and R. E. Schapire, "A Decision-Theoretic Generalization of On-Line Learning and an Application to Boosting," Journal of Computer and System Sciences, vol. 55, pp. 119-139, 1997.
[19] V. N. Vapnik, "The Nature of Statistical Learning Theory," Springer, 1995.
[20] L. Bottou, C. Cortes, J. S. Denker, H. Drucker, I. Guyon, L. D. Jackel, Y. LeCun, U. A. Muller, E. Sackinger, P. Simard, and V. Vapnik, "Comparison of classifier methods: a case study in handwritten digit recognition," in Pattern Recognition - Conference B: Computer Vision & Image Processing., Proceedings of the 12th IAPR International. Conference on, pp. 77-82, 1994.
[21] J. H. Friedman, "Another approach to polychotomous classification," Technical report, Department of Statistics, Stanford University, 1996.
[22] J. C. Platt, N. Cristianini, and J. Shawe-Taylor, "Large margin DAGs for multiclass classification," Advances in Neural Information Processing Systems, vol. 12, pp.547-553, 2000.
[23] D.Klein, "Lagrange Multipliers without Permanent Scarring," Berkeley, 2004.
[24] R. Fletcher, "Practical methods of optimization," 2nd edition, Wiley, 1987.
[25] C. Cortes and V. Vapnik, "Support-vector networks," Machine Learning, vol. 20, pp.273-297, 1995.
[26] R. C. Gonzalez and R. E. Woods, "Digital Image Processing," 2nd edition, Prentice Hall, Inc., 2002.
[27] Chih-Chung Chang and Chih-Jen Lin, LIBSVM : a library for support vector machines, 2001. Software available at http://www.csie.ntu.edu.tw/~cjlin/libsvm