臺灣博碩士論文加值系統

視覺里程計主要用於估算移動物體在運動時，其自身之自運動軌跡。換句話說，視覺里程計就是在進行移動物體的定位。而SLAM系統被認為是視覺領域內空間定位技術的最佳方法。但是SLAM系統相當龐大(前端的視覺里程計，後端的自運動估計誤差之最佳化處理)，即使以桌上型電腦作為SLAM系統的運算平台，也只能算是具有每秒30影格率的處理速度。若是需要同時執行其他的運算處理，在即時性方面將面臨挑戰。因此加強視覺里程計的演算法，在沒有後端的誤差最佳化下，找出可以適當的對移動物體的自運動進行軌跡估計的演算法，即是本論文的研究重點。
在本論文之研究中，主要的貢獻有以下兩點：第一，本論文提出一種名為自運動軌跡估計之影像串列的演算法。透過該演算法的處理，即使不依靠SLAM系統後端之對自運動軌跡估計誤差的最佳化運算，亦能夠適當的對移動物體進行自運動軌跡之估計；第二，本論文所提出之系統在即時性、處理快速、輕巧，以及準確度間能夠取得一個良好的平衡。

Visual odometry is the process of estimating the ego-motion of a moving object. In other words, visual odometry is the process of determining the position of a moving object. Then, the SLAM system is considered to be the best method for spatial positioning technology in the visual field. However, the SLAM system is quite large (the front-end: visual odometry, the back-end: optimization of the ego-motion estimation error), If the system need to perform other arithmetic processing at the same time, it will face challenges in terms of real-time.
There are two contributions of this thesis. First, this thesis proposes an algorithm called image series from ego-motion estimation. Through the processing of the algorithm, even if the optimization of the ego-motion estimation error is not relied on by the back-end of the SLAM system, the estimation of the ego-motion of a moving object can be appropriately performed. Second, the system proposed in this paper can achieve a well balance between real-time, processing speed, lightness, and accuracy.

摘要 ii
ABSTRACT iii
誌謝 iv
目錄 v
圖目錄 vii
表目錄 ix
第一章 緒論 1
1.1 視覺里程計之介紹 1
1.2 研究動機 2
1.3 相關研究之探討 3
1.4 論文貢獻與架構 5
第二章 相關技術與原理 6
2.1 特徵點之偵測 6
2.1.1 Harris角點偵測演算法 7
2.1.2 Shi-Tomasi角點偵測演算法 10
2.2 光流法之特徵點追蹤 11
2.3 多視角幾何學 14
2.3.1 相機矩陣與坐標系 14
2.3.2 對極幾何、基礎矩陣與本質矩陣 21
2.4 求解Perspective n Points問題 26
第三章 系統架構與處理流程 28
3.1 整體系統之架構 29
3.2 特徵點提取 31
3.3 特徵點配對 33
3.3.1 光流法追蹤 34
3.3.2 對極幾何、對極約束 43
3.4 估計特徵點之三維位置 45
3.4.1 自運動軌跡估計之影像串列 46
3.4.2 空間點之重建與融合 51
3.5 估計相機自運動軌跡 58
第四章 系統分析與實驗結果 66
4.1 測試資料介紹 66
4.1.1 第一組測試影像 67
4.1.2 第二組測試影像 68
4.1.3 第三組測試影像 70
4.2 實驗機制說明 71
4.3 系統分析實驗 73
4.3.1 特徵點提取方法之分析與實驗比較 73
4.3.2 特徵點配對方法之分析與實驗比較 79
4.3.3 整體系統演算法對於特徵點處理之測試實驗 83
4.4 系統之評測實驗結果與討論 88
4.4.1 第一組測試影像之實驗結果分析與討論 88
4.4.2 第二組測試影像之實驗結果分析與討論 93
4.4.3 第三組測試影像之實驗結果分析與討論 99
第五章 結論與未來工作 105
參考文獻 106

[1] D. Nister, O. Naroditsky, and J. Bergen, “Visual Odometry,” in Proc. of IEEE Computer Society Conference on Computer Vision and Pattern Recognition, Washington, USA, pp. 652-659, 2004.
[2] R. Mur-Artal, J. M. M. Montiel, and J. D. Tardós, “ORB-SLAM: A Versatile and Accurate Monocular SLAM System,” IEEE Transactions on Robotics, vol. 31, no.5, pp. 1147-1163, 2015.
[3] Raúl Mur-Artal and Juan D. Tardós, “ORB-SLAM2: An Open-Source SLAM System for Monocular, Stereo, and RGB-D Cameras,” IEEE Transactions on Robotics, vol. 33, no.5, pp. 1255-1262, 2017.
[4] J. Engel, T. Schöps, and D. Cremers, “LSD-SLAM: Large-Scale Direct Monocular SLAM,” in European Conference on Computer Vision, Zurich, Switzerland, pp. 834-849, 2014.
[5] C. Harris and M. Stephens, “A Combined Corner and Edge Detector,” in Proc. of 4th Alvey Vision Conference, Manchester, England, vol. 15, pp. 147-151, 1988.
[6] J. Shi and C. Tomasi, “Good features to track,” in Proc. of IEEE Conference on Computer Vision and Pattern Recognition, Seattle, USA, pp. 593-600, 1994.
[7] B. D. Lucas and T. Kanade, “An iterative image registration technique with an application to stereo vision,” in Proc. of the 7th International Joint Conference on Artificial Intelligence, Vancouver, Canada, vol. 2, pp. 674-679, 1981.
[8] R. Hartley and A. Zisserman, Multiple View Geometry in Computer Vision: Second Edition, Cambridge University Press, 2003.
[9] M. A. Fischler and R. C. Bolles, “Random Sample Consensus: A Paradigm for Model Fitting with Applications to Image Analysis and Automated Cartography,” Communications of the ACM, vol. 24, no. 6, pp. 381-395, 1981.
[10] D. Nistér, “An efficient solution to the five-point relative pose problem,” IEEE Transactions on Pattern Analysis and Machine Intelligence. vol. 26, no. 6, pp. 756-770, 2004.
[11] C. Harris and J. Pike, “3d positional integration from image sequences,” Image and Vision Computing, vol. 6, no.2, pp. 87–90, 1988.
[12] W. Forstner, “A feature based correspondence algorithm for image matching,” International Archives of Photogrammetry and Remote Sensing, vol. 26, no. 3, pp. 150-166, 1986.
[13] D. Scaramuzza, F. Fraundorfer, and R. Siegwart, “Real-time monocular visual odometry for on-road vehicles with 1-point RANSAC,” in IEEE International Conference on Robotics and Automation, Kobe, Japan, pp. 4293-4299, 2009.
[14] V. Lepetit, F. Moreno-Noguer, P. Fua, “EPnP: An Accurate O(n) Solution to the PnP Problem,” International Journal of Computer Vision, vol. 81 no. 2, pp. 155-166, 2009.
[15] F. Moreno-Noguer, V. Lepetit, and P. Fua, “Accurate non-iterative O(n) solution to the PnP problem,” in IEEE International Conference on Computer Vision, Rio de Janeiro, Brazil, pp. 1-8, 2007.
[16] L. Kneip, H. Li, Y. Seo, “UPnP: An Optimal O(n) Solution to the Absolute Pose Problem with Universal Applicability,” in European Conference on Computer Vision, Zurich, Switzerland, pp. 127-142, 2014.
[17] H. Badino, A. Yamamoto, and T. Kanade, “Visual odometry by multi-frame feature integration,” in IEEE International Conference on Computer Vision Workshops, Sydney, Australia, pp. 222-229, 2013.
[18] M. Buczko and V. Willert, “Flow-decoupled normalized reprojection error for visual odometry,” in IEEE International Conference on Intelligent Transportation Systems, Rio de Janeiro, Brazil, pp. 1161-1167, 2016.
[19] M. Buczko and V. Willert, “How to distinguish inliers from outliers in visual odometry for high-speed automotive applications,” in IEEE Intelligent Vehicles Symposium, Gothenburg, Sweden, pp. 478-483, 2016.
[20] I. Cvišić and I. Petrović, “Stereo odometry based on careful feature selection and tracking,” in European Conference on Mobile Robots, Lincoln, England, pp. 1-6, 2015.
[21] F. Fraundorfer and D. Scaramuzza, “Visual odometry part II: matching, robustness, optimization, and applications,” IEEE Robotics & Automation Magazine, vol. 19, no. 2, pp. 78-90, 2012.
[22] 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.
[23] A. Geiger, J. Ziegles, and C. Stiller, “StereoScan: Dense 3D Reconstruction in Real-time,” in IEEE Intelligent Vehicles Symposium, Baden-Baden, Germany, pp. 963-968, 2011.
[24] B. Kitt, A. Geiger, and H. Lategahn, “Visual odometry based on stereo image sequences with RANSAC-based outlier rejection scheme,” in IEEE Intelligent Vehicles Symposium, San Diego, USA, pp. 486-492, 2010.
[25] M. Persson, T. Piccini, M. Felsberg, and R. Mester, “Robust stereo visual odometry from monocular techniques,” in IEEE Intelligent Vehicles Symposium, Seoul, South Korea, pp. 686-691, 2015.
[26] I. Cvišić, J. Ćesić, I. Marković and I. Petrović, “SOFT‐SLAM: Computationally efficient stereo visual simultaneous localization and mapping for autonomous unmanned aerial vehicles,” Journal of Field Robotics, vol. 35, no. 4, pp. 578-595, 2018.
[27] H. Badino, “A robust approach for ego-motion estimation using a mobile stereo platform,” in First International Workshop on Complex Motion, Günzburg, Germany, pp. 198-208, 2004.
[28] A. Howard, “Real-time stereo visual odometry for autonomous ground vehicles,” in IEEE/RSJ International Conference on Intelligent Robots and Systems, Nice, France, pp. 3946-3952, 2008.
[29] A. E. Johnson, S. B. Goldberg, Y. Cheng, and L. H. Matthies, “Robust and efficient stereo feature tracking for visual odometry,” in IEEE International Conference on Robotics and Automation, Pasadena, USA, pp. 39-46, 2008.
[30] A. Milella and R. Siegwart, “Stereo-based ego-motion estimation using pixel-tracking and iterative closest point,” in Fourth IEEE International Conference on Computer Vision Systems, New York, USA, pp. 21-21, 2006.
[31] K. Yamaguchi, T. Kato, and Y. Ninomiya, “Vehicle ego-motion estimation and moving object detection using a monocular camera,” in 18th International Conference on Pattern Recognition, Hong Kong, China, pp. 610-613, 2006.
[32] D.G. Lowe, “Object recognition from local scale-invariant features,” in Proc. of the 7th IEEE International Conference on Computer Vision, Kerkyra, Greece, vol. 2, pp. 1150-1157, 1999.
[33] H. Bay, T. Tuytelaars, L. V. Gool, “SURF: Speeded Up Robust Features, ” in European Conference on Computer Vision, Graz, Austria, pp. 404-417, 2006.
[34] E. Rublee, V. Rabaud, K. Konolige, G. Bradski, “ORB: An efficient alternative to SIFT or SURF,” in International Conference on Computer Vision, Barcelona, Spain, pp. 2564-2571, 2011.