系統識別號 | U0002-0109201615154900 |
---|---|
DOI | 10.6846/TKU.2016.00041 |
論文名稱(中文) | 即時RGB-D視覺姿態估測演算法之設計與實現 |
論文名稱(英文) | Design and Implementation of a Real-Time RGB-D Visual Pose Estimation Algorithm |
第三語言論文名稱 | |
校院名稱 | 淡江大學 |
系所名稱(中文) | 電機工程學系碩士班 |
系所名稱(英文) | Department of Electrical and Computer Engineering |
外國學位學校名稱 | |
外國學位學院名稱 | |
外國學位研究所名稱 | |
學年度 | 104 |
學期 | 2 |
出版年 | 105 |
研究生(中文) | 盧家賢 |
研究生(英文) | Chia-Hsien Lu |
學號 | 602470196 |
學位類別 | 碩士 |
語言別 | 繁體中文 |
第二語言別 | |
口試日期 | 2016-07-01 |
論文頁數 | 63頁 |
口試委員 |
指導教授
-
蔡奇謚(chiyi_tsai@mail.tku.edu.tw)
委員 - 翁慶昌(wong@ee.tku.edu.tw) 委員 - 許陳鑑(jhsu@ntnu.edu.tw) |
關鍵字(中) |
視覺姿態估測 RGB-D影像建圖 M型估計式 非線性最佳化 |
關鍵字(英) |
Visual pose estimation RGB-D mapping M-estimator nonlinear optimization |
第三語言關鍵字 | |
學科別分類 | |
中文摘要 |
視覺姿態估測技術為機器人視覺定位系統中一個重要的核心技術,其目的為透過影像特徵點的移動資訊來估測相機本體在空間中的移動資訊。然而,此技術不但運算複雜度高,且容易因錯誤特徵匹配而影響估測準確度。本論文所提出之演算法即為解決使用RGB-D視覺感測資訊來估測相機移動時所面臨的技術問題,並提升估測相機三維旋轉角度及位移姿態之準確性及強健性。透過RGB-D影像中所偵測到的三維特徵匹配點,經由非線性最佳化方式來進行姿態估測運算,來求得相機於空間中的姿態資訊。為了提高系統運算效率,本論文亦經由Jacobian矩陣的整理來降低迭代的複雜度,藉此來加強系統整體的運算速度。本論文另加入M型估計式演算法抑制姿態估測演算法異常值的影響,以得出較穩健的結果。在實驗驗證部分,本論文使用實驗室所拍攝的數據以及Computer Vision Group網站[1]所提供的RGB-D影像的數據,比較三種現有之M型估計式之數學模型,並探討其對結果造成的影響。 |
英文摘要 |
Visual pose estimation technique, which estimates three-dimensional (3D) motion information of a camera system from changes of image features between adjacent frames, is an important core technology in vision-based robot localization systems. However, this technique usually is computationally expensive and is very sensitive to feature matching outliers. To address these technical problems, this thesis presents a RGB-D mapping algorithm that uses RGB-D visual sensing information to improve accuracy and robustness of six Degree-of-Freedom (6 DoF) motion estimation of the camera system. The proposed algorithm estimates the optimal 6 DoF posture information of the camera from the 3D feature matches between two RGB-D frames via a nonlinear optimization process. To improve the computational efficiency of the system, this thesis also derives Jacobian matrix associated with the cost function to reduce computational complexity of the optimization process, thereby enhancing overall system processing speed. Moreover, the proposed algorithm is combined with M-estimators to improve the robustness of the system against the influence of matching outliers. In the experiments, the performance of the proposed algorithm adopting three different types of M-estimators was studied by using RGB-D images corrected in our laboratory and provided on Computer Vision Group website [1]. |
第三語言摘要 | |
論文目次 |
目錄 中文摘要 I 英文摘要 II 目錄 III 圖目錄 V 表目錄 VI 第一章 序論 1 1.1 研究背景 1 1.2 研究動機與目的 4 1.3 論文架構 6 第二章 視覺姿態估測演算法 7 2.1 視覺姿態估測系統架構圖 7 2.2 姿態估測演算法 10 2.3 非線性最佳化演算法 17 2.4 Jacobian矩陣推導 22 2.5 座標轉換 32 第三章 M型估計式演算法 34 3.1 權重函式推導 34 3.2 權重函式之數學模型 36 第四章 實驗結果與分析 41 4.1 軟硬體系統架構 41 4.1.1 硬體介紹 44 4.1.2 RGB-D視覺軟體工具應用 47 4.2 實驗數據 48 4.3 實驗方式 50 4.4 實驗結果 54 4.4.1 Pan-Tilt實驗結果 54 4.4.2 freiburg2/xyz實驗結果 55 第五章 結論與未來展望 58 參考文獻 60 圖目錄 圖2.1、視覺姿態估測演算法示意圖。 9 圖2.2、提出的RGB-D視覺移動估測演算法。 10 圖2.3、非線性最佳化演算法流程圖。 22 圖4.1、軟硬體應用流程圖。 42 圖4.2、Microsoft Kinect RGB-D視覺攝影機。 44 圖4.3、Pan-Tilt Unit-D46控制平台。 46 圖4.4、Pan-Tilt實驗數據圖。 48 圖4.5、freiburg2/xyz實驗數據圖。 49 圖4.6、Pan-Tilt路徑示意圖。 51 圖4.7、freiburg2/xyz實際路徑圖。 53 圖4.8、實驗結果X-Y俯視圖。 57 圖4.9、實驗結果X-Z俯視圖。 57 表目錄 表2.1、數學公式參數化。 24 表2.2、一階微導矩陣。 25 表2.3、對當前x誤差的微分。 26 表2.4、對當前y誤差的微分。 27 表2.5、對當前z誤差的微分。 28 表2.6、對上一刻x誤差的微分。 29 表2.7、對上一刻y誤差的微分。 30 表2.8、對上一刻z誤差的微分。 31 表2.9、對C的微分。 32 表3.1、M型估計式演算法各模型 函式。 37 表3.2、M型估計式演算法各模型 函式和 函式。 37 表3.3、M型估計式演算法各模型模擬圖。 39 表3.4、本研究使用的 函式。 39 表4.1、Microsoft Kinect相關規格表。 45 表4.2、Pan-Tilt Unit-D46相關規格表。 46 表4.3、電腦規格表。 47 表4.4、Pan-Tilt實驗結果表。 55 表4.5、Pan-Tilt實際角度[0.0,0.0,0.0]座標表。 55 表4.6、Pan-Tilt實際位移[0.0,0.0,0.0]座標表。 55 表4.7、freiburg2/xyz位移實驗結果表。 56 表4.8、freiburg2/xyz角度實驗結果表。 56 |
參考文獻 |
[1] https://vision.in.tum.de/data/datasets/rgbd-dataset/download [2] D. Nister, “An efficient solution to the five-point relative pose problem,” IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 26, No. 6, 2004, pp. 756-770. [3] A. Howard, “Real-time stereo visual odometry for autonomous ground vehicles,” IEEE/RSJ International Conference on Intelligent Robots and Systems, 2008, pp. 3946-3952. [4] J. R. Fabian, G. M. Clayton, “Adaptive visual odometry using RGB-D cameras,” IEEE/ASME International Conference on Advanced Intelligent Mechatronics, 2014, pp. 1533-1538. [5] https://en.wikipedia.org/wiki/Kinect#Kinect_for_Windows [6] I. Dryanovski, R. G. Valenti, J. Xiao, “A fast visual odometry and mapping system for RGB-D cameras,” IEEE International Conference on Robotics and Automation (ICRA), 2013, pp. 2305-2310. [7] S. Manoj Prakhya, L. Bingbing, L. Weisi, U. Qayyum, “Sparse depth odometry : 3D keypoint based pose estimation from dense depth data,” IEEE International Conference on Robotics and Automation (ICRA), 2015, pp. 4216-4223. [8] J. Helge Klüssendorff, J. Hartmann, D. Forouher, E. Maehle, “Graph-based visual SLAM and visual odometry using an RGB-D camera,” 9th Workshop on Robot Motion and Control (RoMoCo), 2013, pp. 288-293. [9] D. Lowe, “Object recognition from local scale-invariant features,” in Proceedings of the Seventh IEEE International Conference on Computer Vision (ICCV), 1999, pp. 1150-1157. [10] H. Bay, A. Ess, T. Tuytelaars, and L. Van Gool, “Speeded-up robust features (SURF),” Computer Vision and Image Understanding, Vol. 110, No. 3, 2008, pp. 346-359. [11] M. Nowicki, P. Skrzypezyński, “Combining photometric and depth data for lightweight and robust visual odometry,” European Conference on Mobile Robots (ECMR), 2013, pp. 125-130. [12] B. Marques Ferreira da Silva, L. Marcos Garcia Gonçalves, “A fast feature tracking algorithm for visual odometry and mapping based on RGB-D sensors,” 27th SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI), 2014, pp. 227-234. [13] M. Ozuysal, M. Calonder, V. Lepetit, and P. Fua, “Fast keypoint recognition using random ferns.” IEEE Transactions on Pattern Analysis and Machine Intelligence,Vol. 32, No. 3, 2010, pp. 448-461. [14] F. Steinbrücker, J. Sturm, D. Cremers, “Real-time visual odometry from dense RGB-D images,” IEEE International Conference on Computer Vision Workshops (ICCV Workshops), 2011, pp. 719-722. [15] Ying-Hao Wang, Hao-En Cheng, Chih-Jui Lin, Ri-Wei Deng, Hsuan Lee, Tzuu-Hseng S. Li, “Realization of affine SIFT real-time image processing for home service robot,” International Conference on System Science and Engineering (ICSSE), 2013, pp. 141-146. [16] Eric Li, Liu Yang, Bin Wang, Jianguo Li, Ya-ti Peng, “SURF cascade face detection acceleration on sandy bridge processor,” IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, 2012, pp.41-47. [17] S. Klose, P. Heise, A. Knoll, “Efficient compositional approaches for real-time robust direct visual odometry from RGB-D data,” IEEE/RSJ International Conference on Intelligent Robots and Systems, 2013. , pp. 1100-1106 [18] K. G. Derpanis, “Overview of the RANSAC algorithm,” Technical report, Computer Science, York University, 2010. [19] G. Hee Lee, F. Fraundorfer, M. Pollefeys, “RS-SLAM RANSAC sampling for visual FastSLAM,” IEEE/RSJ International Conference on Intelligent Robots and Systems, 2011, pp. 1655-1660. [20] Z. Zhang, “Parameter estimation techniques: a tutorial with application to conic fitting,” Image and Vision Computing, Vol.15, No.1, 1997, pages 59-76. [21] AMC, “Robust statistics: a method of coping with outliers,” Royal Society of Chemistry, 2001. [22] C. Chen, “Robust regression and outlier detection with the ROBUSTREG procedure,” Proceedings of the Twenty-seventh Annual SAS Users Group International Conference, 2002. [23] P. J. Rousseeuw, C. Croux, “Alternatives to the median absolute deviation,” Journal of the American Statistical Association, Vol. 88, No. 424, 2012, pp. 1273-1283. [24] R. Chartrand, W. Yin, “Iteratively reweighted algorithms for compressive sensing,” IEEE International Conference on Acoustics, Speech and Signal Processing, 2008, pp. 3869-3872. [25] T. Tykkälä, C. Audras, A. I. Comport, “Direct iterative closest point for real-time visual odometry,” IEEE International Conference on Computer Vision Workshops (ICCV Workshops), 2011, pp. 2050-2056. [26] N. Karlsson, E. di Bernardo, J. Ostrowski, L. Goncalves, P. Pirjanian, M. E. Munich, “The vSLAM algorithm for robust localization and mapping,” Proceedings of the 2005 IEEE International Conference on Robotics and Automation, 2005, pp. 24-29. [27] R. G. Valenti, I. Dryanovski, C. Jaramillo, D. Perea Ström, J. Xiao, “Autonomous quadrotor flight using onboard RGB-D visual odometry,” IEEE International Conference on Robotics and Automation (ICRA), 2014, pp. 5233-5238. [28] D. Li, Q. Li, N. Cheng, Q. Wu, J. Song, L. Tang, “Combined RGBD-inertial based state estimation for MAV in GPS-denied indoor environments,” 9th Asian Control Conference (ASCC), 2013, pp. 1-8. [29] Z. Fang, S. Scherer, “Experimental study of odometry estimation methods using RGB-D cameras,” IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2014. [30] 黃志弘,立體視覺里程計演算法之設計與實現,淡江大學電機工程學系碩士論文(指導教授:蔡奇謚),2013。 |
論文全文使用權限 |
如有問題,歡迎洽詢!
圖書館數位資訊組 (02)2621-5656 轉 2487 或 來信