臺灣博碩士論文加值系統

交通事故往往造成人員傷亡及財產損失，而許多交通事故的發生皆因駕駛疏失使得車輛偏離車道而造成意外。近年來政府及汽車產業無不持續提出車道偏移警示系統(Lane Departure Warning System, LDWS)的研究，希望能減少交通事故的發生。本論文運用電腦視覺(Computer vision)技術快速偵測直線或彎道的車道標線(Lane marking)。並記錄過去車輛行駛路徑與車道標線之間的偏移量，進而預測未來虛擬行駛路徑，用以判斷車輛是否偏離車道，藉此對車輛駕駛提出警示。
本系統首先利用車道標線的二次曲線參數使用卡爾曼濾波器(Kalman filter)預測更新車道標線，建立感興趣區域(Region of Interest, ROI)。影像前處理是使用K-means分群演算法(K-means clustering)進行影像二值化(Binarization)，再將二值化影像以Sobel運算(Sobel operator)提取車道標線邊緣影像，並且在車道標線邊緣影像中使用影像形態學(Morphology)方法消除影像中的微小雜訊，以及增強、修補車道標線的特徵。接著將ROI區域配合霍夫轉換(Hough transformation)偵測經影像前處理後影像中的車道標線，將所偵測車道標線資訊經由最小平方法(Least square)繪出影像中車道標線位置。最後車道偏離預警是運用灰色理論(Grey theory)將記錄的車道標線資料及駕駛路徑，用來預測未來虛擬駕駛行駛路徑。
由實驗結果可以看出，本論文所預測的未來虛擬駕駛行駛路徑可進而計算穿越車道標線的時間(Time to Lane Crossing, TLC)與穿越車道標線的距離(Distance to Lane Crossing, DLC)。經由預測所得的TLC及DLC，藉以判斷車輛未來是否會偏離駕駛車道，可以事先警告駕駛者是否即將偏離車道、注意行車安全，以減少事故發生，此虛擬行駛路徑是經由記錄過去的行駛路徑預測所得，因此系統具有適應性。

Car accidents cause not just financial damage but physical damage as well. Many accidents occur as a result of negligent driving, by drivers who drift outside of the marked lanes. Recently, research has been done on Lane Departure Warning (LDW) systems with the aim of minimizing car accidents. This research using computer vision technology detects the lane markings, and to record the amount of deviation between the past vehicle route and lane marking, which are used to project future virtual lanes and then predict if the car is going to deviate from these lanes.
This research using the Kalman filter to predict the updated lane marking, and to create the Region of Interest. Image preprocessing uses the k-means clustering method to proceed on binarization of the image and the Sobel operator to extract the edges of the images of the lane markings. Using morphology to reduce the noise in the marginal image, and enhance and repair lane marking. The ROI combined with Hough transform detect the lane markings from the preprocessed image, and using a least squares fit, the location of the lane markings is obtained. Finally, the grey theory is applied to predict the future of the vehicle virtual path.
As the result of the experiment, it has clearly shown that the prediction of the vehicle route can further estimate the time to lane crossing (TLC), and the distance to lane crossing (DLC). By using TLC and DLC, it can help drivers to make sure they are driving in the correct lane, and prevent them from traffic accidents. The system is adaptive since the virtual path estimation is based on past driving paths.

摘要 i
Abstract ii
誌謝 iii
目錄 iv
圖目錄 vi
表目錄 viii
第一章 緒論 1
1.1 研究背景 1
1.2 研究動機與目的 4
1.3 系統架構 5
1.4 論文貢獻 8
1.5 論文架構 9
第二章 相關研究 10
2.1 概述 10
2.2 車道標線偵測 10
2.3 車道偏離預警 13
第三章 車道標線偵測 17
3.1 概述 17
3.2 縮小車道標線搜尋範圍 18
3.2.1 卡爾曼濾波器 18
3.2.2 初始感興趣區域建立 22
3.2.3 追蹤感興趣區域建立 28
3.3 強化車道標線特徵 28
3.3.1 增強黃色車道標線 29
3.3.2 色彩空間轉換 30
3.3.3 影像二值化 32
3.3.4 邊緣提取 35
3.3.5 增強車道標線特徵 37
3.4 車道標線特徵擷取 40
3.5 標定車道標線位置 44
第四章 車道偏離預警 47
4.1 概述 47
4.2 車道偏移量計算 48
4.3 未來虛擬駕駛路徑預測 51
第五章 實驗結果與分析 54
5.1 實驗方法說明 54
5.2 黃色車道標線增強與沒增強結果比對 56
5.3 特殊的車道標線偵測結果 57
5.4 開發平台車道標線偵測結果 60
5.5 車道標線追蹤偵測結果 61
5.6 未來虛擬路徑預測結果 65
5.7 結果探討 69
第六章 結論 71
6.1 結論 71
6.2 未來展望 72
參考文獻 73

[1] 行政院衛生署,http://www.doh.gov.tw/CHT2006/index_populace.aspx.
[2] 交通部高公局,http://www.freeway.gov.tw/Publish.aspx?cnid=516&;p=2849.
[3] TRI拓墣產業研究所，智慧型車輛引爆車用電子新發展，TRI產業專題報告，2008。
[4] Jyun-Guo Wang, Cheng-Jian Lin, Shyi-Ming Chen, “Applying fuzzy method to vision-based lane detection and departure warning system”, Expert Systems with Applications, Vol. 37, pp. 113-126, 2010.
[5] Stephen P. Tseng, Yung-Sheng Liao, Chih-Hsie Yeh, et al, “A DSP based Lane Recognition Method for the Lane Departure Warning System of Smart Vehicles”, IEEE International Conference on Networking, Sensing and Control, pp. 823-828, 2009.
[6] 張為彥，“基於模糊演算法之車道偏離預警系統” ，國立勤益科技大學電機工程系，碩士論文，2011。
[7] 楊丰盛，“android應用開發揭秘” ，佳魁資訊股份有限公司， 2010。
[8] 陳璿文，“以視覺為基礎之全日嵌入式車輛偵測系統”，國立勤益科技大學電機工程系，碩士論文，2010。
[9] Jae Kyu Suhr, Ho Gi Jung, “Fully-automatic Recognition of Various Parking Slot Markings in Around View Monitor (AVM) Image Sequences”, IEEE International Conference on Intelligent Transportation Systems, pp. 1294-1299, 2012.
[10] Hendrik Weigel, Philipp Lindner, Gerd Wanielik, “Vehicle Tracking with Lane Assignment by Camera and Lidar Sensor Fusion”, IEEE Intelligent Vehicles Symposium, pp. 513-520, 2009.
[11] 徐溫嶺，“應用於智慧型車輛之駕駛疲勞偵測系統”，國立勤益科技大學電機工程系，碩士論文，2010。
[12] Bin-Feng Lin, Yi-Ming Chan, Li-Chen Fu, et al, “Incorporating Appearance and Edge Features for Vehicle Detection in the Blind-spot Area”, IEEE International Conference on Intelligent Transportation Systems, pp. 869-874, 2010.
[13] 顏志鴻，“使用Fisher線性判別分析之快速行人偵測系統”，國立勤益科技大學電機工程系，碩士論文，2011。
[14] Yue Wang, Eam Khwang Teoh, Dinggang Shen, “Lane detection and tracking using B-Snake”, powered by Science Direct Image and Vision Computing, Vol. 22, pp. 269-280, 2004.
[15] Shengyan Zhou, Yanhua Jiang, Junqiang Xi, et al, “A Novel Lane Detection based on Geometrical Model and Gabor Filter”, IEEE Intelligent Vehicles Symposium, pp. 59-64, 2010.
[16] Kiyokazu Takagi, Katsuhiro Morikawa, Takashi Ogawa, et al, “Road Environment Recognition Using On-vehicle LIDAR”, IEEE Intelligent Vehicles Symposium, pp. 120-125, 2006.
[17] Guoliang Liu, Florentin W¨org¨otter, Irene Markeli´c, “Combining Statistical Hough Transform and Particle Filter for Robust Lane Detection and Tracking”, Intelligent Vehicles Symposium, pp. 993-997, 2010.
[18] Qing Lin, Youngjoon Han, Hernsoo Hahn, “Real-time Lane Detection Based on Extended Edge-linking Algorithm”, IEEE Second International Conference on Computer Research and Development, pp. 725-730, 2010.
[19] Pei-Yung Hsiao, Kuo-Chen Hung, Shih-Shinh Huang, et al, “An embedded lane departure warning system”, IEEE International Symposium on Consumer Electronics, pp. 162-165, 2011.
[20] Yu-Chi Leng, Chieh-Li Chen, “Vision-Based Lane Departure Detection System in Urban Traffic Scenes”, IEEE International Conference on Control Automation Robotics &; Vision, pp. 1875-1880, 2010.
[21] R.E. Kalman, “A New Approach to Linear Filtering and Prediction Problems”, Transaction of the ASME-Journal of Basic Engineering, Vol. 82, no. 1, pp. 34-45, 1960.
[22] Roland Chapuis, Romuald Aufrere, Frédéric Chausse, “Accurate Road Following and Reconstruction by Computer Vision”, IEEE Transactions on Intelligent Transportation Systems, Vol. 3, NO. 4, pp. 261-270, 2002.
[23] HunterLab, "CIE L*a*b* Color Scale", Technical Services Department Hunter Associates Laboratory, Vol.8, N0.7, pp.1-4, 2008.
[24] Sharma, Subhash, “Applied Multivariate Techniques”, John Wiley &; Sons Inc, 1996.
[25] V. Leavers, “Which Hough Transform？”, CVGIP: Image Understanding, Vol. 58, No. 2, pp. 250-264, 1993.
[26] 李彥鋒，“以視覺為基礎之嵌入式車輛偵測系統”，國立勤益科技大學電機工程系，碩士論文，2009。
[27] D. Julong, “Control problems of grey system,” System and Control Letters, pp. 288-294, 1982.
[28] 交通部台灣區高速公路局，道路交通標誌標線號誌設置規則，2008年。