臺灣博碩士論文加值系統

本文為實現先進駕駛輔助系統(Advanced Driver Assistance Systems, ADAS)晶片化，達到系統落地之目的，設計過程將透過單鏡頭攝影機執行影像處理演算法，並設計基於汽車行駛規則之專家系統，減少運算成本與記憶體消耗，克服在目前多數研究中，以類神經網路與機械學習作為演算法而無法在本地端執行之問題。
本文以單鏡頭攝影機基於光流法做為系統核心演算法，藉由偵測出移動視覺擴張焦點(Focus of Expansion, FoE)，建構影像視覺深度訊息，實現智慧車輛視覺輔助系統，其中系統架構包含：智慧自適應車燈(Intelligent Adapted Headlamp, IAH)、道路偏移警示系統(Lane Departure Warning System, LDWS)和同步定位與地圖構建系統(Simultaneous localization and mapping, SLAM)，為智能車搭建視覺感官，接收周邊行駛環境資訊，避免出現危險駕駛行為及碰撞障礙物的可能。
研究中所測試的影像資料集皆出至於德國卡爾斯魯厄理工學院和豐田美國技術研究院(Karlsruhe Institute of Technology and Toyota Technological Institute, KITTI)聯合創辦的計算機視覺演算法評測資料庫，此組織收納各種行車環境可讓系統驗證，本文將擷取市區道路、州際公路、鄉間小路三種環境做驗證，證明研究內容可行性。

In order to transform the Advanced Driver Assistance Systems (ADAS) into a system chip for the goal of Real-time computing. To solve the problem that Neural Network and Machine Learning cannot execute locally with real-time at most of ADAS studies, this thesis will execute image processing through a monocular camera and design an expert system based on car driving rules to reduce computing costs and memory consumption.
In this thesis, the image processing system is based on the optical flow as the core algorithm by the monocular camera. By detecting the Focus of Expansion (FoE) the depth information of the image vision is constructed to realize the intelligent vehicle visual assistance system. The system architecture includes：Intelligent Adapted Headlamp (IAH), Lane Departure Warning System (LDWS) and Simultaneous Localization and Mapping (SLAM), to build self-driving car with visual senses and receive information on the surrounding driving environment to avoid dangerous driving behavior and the possibility of collision with obstacles.
The datasets are all from the Karlsruhe Institute of Technology in Germany and Toyota Technological Institute (KITTI). We will extract three environments for verification：Suburban, urban and interstate highway, to prove the feasibility of this system.

摘要 i
ABSTRACT ii
致謝 iii
目錄 iv
表目錄 vi
圖目錄 vii
縮寫列表 x
符號說明 xi
壹、緒論 1
一、簡介 1
二、貢獻 2
三、引文 4
貳、研究與方法 7
一、智慧自適應車燈 (Intelligent Adaptive Headlight, IAH) 7
(一).路面區域及ROI判定區域設定 8
(二).粗像素分割 9
(三).判定機制 10
二、流場建立 11
(一).緊密流場－Farnebäck Optical Flow Algorithm 13
(二).稀疏流場 － Harris Orb Matching Flow 14
(三).緊密流場過濾器 15
(四).稀疏流場過濾器 16
三、移動視覺擴張焦點(Focus of Expansion, FoE)偵測 17
(一).收集FoE運算因子 18
(二).FoESreg、FoEDreg候選點偵測 19
(三).FoEDreg 、 FoESreg 過濾器 20
(四).權重投票箱統計與迴授 21
(五).特殊情況 22
四、視線消失點(Vanish Point)偵測 23
(一). Canny 邊緣偵測 24
(二). 霍夫轉換(Hough Transform) 24
(三). 視線消失點偵測(Vanish Point Detector) 25
(四). 投票箱統計與迴授 25
五、同步定位與地圖構建系統
(Simultaneous Localization and Mapping , SLAM) 26
(一). 幾何距離地圖建立 26
(二). 車體運動姿態偵測 31
參、實驗成果展示 44
一、智慧自適應車燈IAH 44
(一). 系統反饋 44
(二). 實際場域測試 46
(三). IAH系統效能 47
二、同步定位與地圖構建SLAM 47
(一). 道路偏移警示系統 48
(二). 深度偵測 52
(三). 姿態偵測 53
陸、結論 58
一、研究總結 58
二、未來展望 59
柒、參考文獻 60

[1]許智偉, "基於地標以及路面偵測之先進駕駛輔助系統晶片雛形實現," 碩士, 電子工程系碩士班, 國立高雄第一科技大學, 高雄市, 2018. [Online]. Available: https://hdl.handle.net/11296/s9x828
[2]S. Jin et al., "FPGA design and implementation of a real-time stereo vision system," IEEE transactions on circuits and systems for video technology, vol. 20, no. 1, pp. 15-26, 2009.
[3]B. K. Horn, "Closed-form solution of absolute orientation using unit quaternions," Josa a, vol. 4, no. 4, pp. 629-642, 1987.
[4]S. Li, C. Xu, and M. Xie, "A robust O (n) solution to the perspective-n-point problem," IEEE transactions on pattern analysis and machine intelligence, vol. 34, no. 7, pp. 1444-1450, 2012.
[5]M. Montemerlo, S. Thrun, D. Koller, and B. Wegbreit, "FastSLAM: A factored solution to the simultaneous localization and mapping problem," Aaai/iaai, vol. 593598, 2002.
[6]A. J. Davison, I. D. Reid, N. D. Molton, and O. Stasse, "MonoSLAM: Real-time single camera SLAM," IEEE transactions on pattern analysis and machine intelligence, vol. 29, no. 6, pp. 1052-1067, 2007.
[7]R. Mur-Artal, J. M. M. Montiel, and J. D. Tardos, "ORB-SLAM: a versatile and accurate monocular SLAM system," IEEE transactions on robotics, vol. 31, no. 5, pp. 1147-1163, 2015.
[8]R. Jain, "Direct computation of the focus of expansion," IEEE Transactions on Pattern Analysis and Machine Intelligence, no. 1, pp. 58-64, 1983.
[9]K. Suzuki and H. Jansson, "An analysis of driver’s steering behaviour during auditory or haptic warnings for the designing of lane departure warning system," JSAE review, vol. 24, no. 1, pp. 65-70, 2003.
[10] C.-K. Yao, Y.-R. Lin, Y.-F. Su, and N.-S. Chang, "VLSI implementation of a real-time vision based lane departure warning system," in 2012 12th International Conference on ITS Telecommunications, 2012: IEEE, pp. 170-174.
[11]J. Son, H. Yoo, S. Kim, and K. Sohn, "Real-time illumination invariant lane detection for lane departure warning system," Expert Systems with Applications, vol. 42, no. 4, pp. 1816-1824, 2015.
[12]P. Baraldi, E. D. Micheli, and S. Uras, "Motion and Depth from Optical Flow," presented at the Procedings of the Alvey Vision Conference 1989, 1989.
[13] Y. Chen, M. Alain, and A. Smolic, "Fast and accurate optical flow based depth map estimation from light fields," in Irish Machine Vision and Image Processing Conference (IMVIP), 2017.
[14] H. Kong, J.-Y. Audibert, and J. Ponce, "Vanishing point detection for road detection," in 2009 IEEE Conference on Computer Vision and Pattern Recognition, 2009: IEEE, pp. 96-103.
[15]H. T. Nguyen, M. Worring, and R. Van Den Boomgaard, "Watersnakes: Energy-driven watershed segmentation," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 25, no. 3, pp. 330-342, 2003.
[16] G. Farnebäck, "Two-frame motion estimation based on polynomial expansion," in Scandinavian conference on Image analysis, 2003: Springer, pp. 363-370.
[17]K. G. Derpanis, "The harris corner detector," York University, pp. 1-2, 2004.
[18]M. Calonder, V. Lepetit, M. Ozuysal, T. Trzcinski, C. Strecha, and P. Fua, "BRIEF: Computing a local binary descriptor very fast," IEEE transactions on pattern analysis and machine intelligence, vol. 34, no. 7, pp. 1281-1298, 2011.
[19]E. Karami, S. Prasad, and M. Shehata, "Image matching using SIFT, SURF, BRIEF and ORB: performance comparison for distorted images," arXiv preprint arXiv:1710.02726, 2017.
[20]G. Welch and G. Bishop, "An introduction to the Kalman filter," ed: Citeseer, 1995.
[21]P. Bao, L. Zhang, and X. Wu, "Canny edge detection enhancement by scale multiplication," IEEE transactions on pattern analysis and machine intelligence, vol. 27, no. 9, pp. 1485-1490, 2005.
[22]D. H. Ballard, "Generalizing the Hough transform to detect arbitrary shapes," in Readings in computer vision: Elsevier, 1987, pp. 714-725.
[23]曾清红 and 卢德唐, "基于移动最小二乘法的曲线曲面拟合," 工程圖學學報, vol. 25, no. 1, pp. 84-89, 2004.
[24]A. Geiger, P. Lenz, C. Stiller, and R. Urtasun, "Vision meets robotics: The kitti dataset," The International Journal of Robotics Research, vol. 32, no. 11, pp. 1231-1237, 2013.