臺灣博碩士論文加值系統

近幾年來，智慧型運輸系統 (Intelligent Transportation System, ITS) 正快速地發展，而先進駕駛輔助系統 (Advanced Driver Assistance Systems) 能夠在駕駛環境中適時警告駕駛者其潛在危險性，更是其中最重要的一環。在國外已有 VOLVO 與 Mercedes-Benz，而國內則有裕隆汽車等大廠投入影像式感測器於行車安全的研究，顯示出影像式系統在行車安全之應用尉為主流。
本論文提出一基於立體視覺影像分析之先進汽車駕駛安全輔助技術。其主要的設計精神係透過立體視差計算、U-V 視差分析以及全局運動估計等技術發展出『疑似障礙物特徵偵測模組』、『三維空間偵測模組』、『疑似障礙物行為推論模組』、『安全警示模組』，並將這些模組整合出先進汽車駕駛安全輔助系統。
根據 C 語言的模擬實驗結果顯示，本論文提出的基於立體視覺影像分析之先進汽車駕駛安全輔助技術在測試的九組影像序列中平均F-measure 為 84.9 %，且能夠在具潛在危險性的場景中適時對駕駛人提出警示。

誌 謝 i
中文摘要 ii
目錄 iii
圖目錄 v
表目錄 vii
第一章 簡介 1
1.1 動機 1
1.2 文獻綜覽 1
1.3 研究方法 8
1.4 論文架構 8
第二章 先進汽車駕駛安全輔助系統架構 9
2.1 系統概述 9
2.2 疑似障礙物特徵偵測模組 11
2.3 三維空間偵測模組 15
2.4 疑似障礙物行為推論模組 30
2.5 安全警示模組 34
第三章 立體視覺匹配演算法 36
3.1 立體視覺計算 36
3.2 子視窗設定 37
3.3 循環成本累積 38
3.4 不可靠視差值偵測 39
3.5 針對車用立體影像之相關改良 39
3.5.1 子視窗二次擴展 39
3.5.2 基於梯度之視差成本函數分析 41
3.5.3 可適應性支持子視窗上下擴展限制 42
3.5.4 視差圖後處理 43
第四章 實驗結果與分析 46
4.1 深度估測比較 47
4.2 障礙物偵測實驗結果 48
4.3 與文獻 [16] 偵測結果比較 55
4.4 本系統偵測效能評估 57
第五章 結論與未來工作 61
5.1 結論 61
5.2 未來工作 61
參考文獻 62

[1]Zhou Junjing, Duan Jianmin, and Yu Hongxiao, “Machine-vision based preceding vehicle detection algorithm: A review,” Proc. Of IEEE the 10th World Congress on Intelligent Control and Automation (WCICA), Beijing, China, July 2012.
[2]Zehang Sun, George Bebis, and Ronald Miller, “On-road vehicle detection: a review,” IEEE Trans. on Pattern Analysis and Machine Intelligence (TPAMI), Vol. 28, No. 5, pp. 694-711, May 2006.
[3]D. Geronimo, A.M. Lopez, A.D. Sappa, and T. Graf, “Survey of Pedestrian Detection for Advanced Driver Assistance Systems,” IEEE Trans. on Pattern Analysis and Machine Intelligence (TPAMI), Vol. 32, No.7, pp. 1239-1258, July 2010.
[4]Sayanan Sivaraman, and Mohan M. Trivedi, “Combining monocular and stereo-vision for real-time vehicle ranging and tracking on multilane highways,” Proc. Of IEEE Int’l Conf. On Intelligent Transportation Systems (ITSC), Washington, DC, USA, Oct. 2011.
[5]A. Leu, D. Aiteanu, and A. Graser, “A novel stereo camera based collision warning system for automotive applications,” Proc. Of IEEE the 6th Int’l Symposium on Applied Computational Intelligence and Informatics (SACI), Timisoara, Romania, May 2011.
[6]Raphael Labayrade, Didier Aubert, and Jean-Philippe, “Real time obstacle detection in stereovision on non flat road geometry through "V-disparity" representation,” Proc. Of IEEE Int’l Conf. On Intelligent Vehicle Symposium (IV), Vol. 2, pp.646-651, June 2002.
[7]Yeping Su, Ming-Ting Sun, and Vincent Hsu, “Global motion estimation from coarsely sampled motion vector field and the applications,” IEEE Trans. on Circuits and Systems for Video Technology, Vol. 15, No. 2, pp.232-242, Feb. 2005.
[8]“KLT : An Implementation of the Kanade-Lucas-Tomasi Feature Tracker” at http://www.ces.clemson.edu/~stb/klt/
[9]楊沅智，可適應天候變化的前車碰撞警示系統，國立中央大學資訊工程研究所碩士論文，中華民國一百零一年七月。
[10]鄭安傑，以電腦視覺分析為基礎的車輛行駛安全防護技術及在 TiDaVinci 雙核心平台上之實現，中華民國九十七年七月。
[11]R. C. Gonzalez, and R. E. Woods, “Digital Image Processing.” 3rd Ed.,Prentice Hall, Upper Saddle River, NJ, 2002
[12]Ta-Te Lin, An-Chih Tsai, Kai-Chiang Chuang, Yu-Chou Chen, and Yu-Sung Chen, “A Real-Time Stereo Vision System for Obstacle Detection and Recognition,” Proc. Of American Society of Agricultural and Biological Engineers (ASABE), Dallas, Texas, August 2012.
[13]陳育菘、楊志弘，基於立體視覺之前方安全警示技術開發，中華民國第十七屆車輛工程學術研討會，中華民國九十九年十一月。
[14]Zhencheng Hu, Francisco Lamosa, and Keiichi Uchimura, “A Complete U-V-disparity Study for Stereovision Based 3D Driving Environment Analysis,” Proc. Of IEEE Int’l Conf. On 3-D Digital Imaging and Modeling (3DIM), pp.204-211, June 2005.
[15]陳軍宇，基於適應性子視窗之循環累積相似度量測的平行化立體視覺匹配演算法，國立中正大學電機工程研究所碩士論文，中華民國一百年七月。
[16]Chung-Hee Lee, Young-Chul Lim, Soon Kwon, and Jonghwan Kim, “Stereo vision-based obstacle detection using dense disparity map,” Proc. Of SPIE Int’l Conf. On Graphic and Image Processing (ICGIP), Cairo, Egypt, Oct. 2011.
[17]Y. Gao, X. Ai, Y. Wang, J. Rarity, and N. Dahnoun, “U-V-Disparity based Obstacle Detection with 3D Camera and steerable filter,” Proc. Of IEEE Int’l Conf. On Intelligent Vehicle Symposium (IV), Baden-Baden, Germany, June 2011.
[18]Intae Na, Sang Hyun Han, and Hong Jeong, “Stereo-based road obstacle detection and tracking,” Proc. Of IEEE Int’l Conf. On Advanced Communication Technology (ICACT), Seoul, Korea, Feb. 2011.
[19]Sebastien Kramm, and Abdelaziz Bensrhair, “Obstacle detection using sparse stereovision and clustering techniques,” Proc. Of IEEE Int’l Conf. On Intelligent Vehicle Symposium (IV), Alcala de Henares, Spain, June 2012.
[20]Vinh Dinh Nguyen, Thuy Tuong Nguyen, Dung Duc Nguyen, and Jae Wook Jeon, “Toward Real-Time Vehicle Detection Using Stereo Vision and an Evolutionary Algorithm,” Proc. Of IEEE Int’l Conf. On Vehicular Technology Conference (VTC Spring), Yokohama, Japan, May 2012.
[21]S. Lefebvre, and S. Ambellouis, “Vehicle detection and tracking using Mean Shift segmentation on semi-dense disparity maps,” Proc. Of IEEE Int’l Conf. On Intelligent Vehicle Symposium (IV), Alcala de Henares, Spain, June 2012.
[22]Truman Shen, Greg Schamp, Troy Cooprider, and Faroog Ibrahim, “Stereo vision based full-range object detection and tracking,” Proc. Of IEEE Int’l Conf. On Intelligent Transportation Systems (ITSC), Washington, DC, USA, Oct. 2011.
[23]P. Lenz, J. Ziegler, A. Geiger, and M. Roser, “Sparse scene flow segmentation for moving object detection in urban environments,” Proc. Of IEEE Int’l Conf. On Intelligent Vehicle Symposium (IV), Baden-Baden, Germany, June 2011.
[24]N. Ben Romdhane, M. Hammami, and H. Ben-Abdallah, “A generic obstacle detection method for collision avoidance.” Proc. Of IEEE Int’l Conf. On Intelligent Vehicle Symposium (IV), Baden-Baden, Germany, June 2011.
[25]Tang-Hsien Chang, and Chen-Ju Chou, "Rear-end collision warning system on account of a rear-end monitoring camera" in Proc. IEEE Conf. Intelligent Vehicles Symposium (IV), Xi'an, June 3-5, 2009, pp.913-917.
[26]"the KITTI Vision Benchmark Suite" at http://www.cvlibs.net/datasets/kitti/