臺灣博碩士論文加值系統

本論文專注於研究聚合不同變焦影像資訊之立體匹配技術。在此系統中，透過變焦相機取得不同焦距下之影像，藉由聚合不同變焦影像資訊，進而改善原有的區域及全域立體匹配技術之視差影像。此外，在論文當中提出變焦矯正技術，透過此技術，變焦影像之間可以藉由變焦向量正確的搜尋到對應點。在實驗當中，除了對實際影像所產生之視差影像做錯誤匹配率的比較，也利用了 Middlebury 提供的測試資料，透過數位變焦的方式得到不同變焦影像，並利用此影像對本論文所提出的方法進行實驗。最後從實驗結果發現，區域性及全域性立體視覺演算法搭配本篇論文提出的架構能成功降低錯誤匹配率。

This thesis investigates stereo matching with zooming information.In the proposed technique, two zoom lens cameras are used to capture multiple stereo image pairs with different focal length.With the assistance of zoom image pairs captured from the same camera, it is possible to increase the correctness of stereo matching results.In addition, a zoom rectification method is proposed to simplify the zoom vector computation.In the experiments, we use the bad pixel rate to compare the results obtained from the real scene images using conventional stereo matching techniques and the proposed stereo with zooming approach. Moreover, we test the proposed algorithm on the Middlebury dataset with digital zoom to demonstrate the effectiveness of our work.It is shown that our technique is able to reduce the bad pixel rate using both the local and global stereo matching algorithms.

摘要 i
Abstract ii
誌謝 iii
圖目錄 ix
表目錄 x
中英文字對照 xi
1 緒論 1
1.1 研究動機 1
1.2 論文架構 2
2 背景知識 3
2.1 變焦影像 3
2.2 立體視覺理論 4
2.2.1 立體視覺影像匹配 5
2.2.2 同軸幾何 9
2.2.3 影像校正與立體視覺系統 10
2.3 單應性矩陣 15
3 基於變焦影像資訊整合之立體視覺 17
3.1 系統概述 17
3.2 影像前處理 19
3.2.1 特徵點擷取與匹配 19
3.2.2 變焦矯正 21
3.2.3 基於特徵點找尋影像中心 25
3.3 匹配代價計算 27
3.3.1 左右視差影像雙向校對 27
3.3.2 基於變焦向量之對應方法 29
3.3.3 基於變焦向量匹配點代價之聚合及視差選取 31
4 系統整合與實驗結果 34
4.1 硬體設備與實驗環境 34
4.2 Middlebury dataset 立體匹配 35
4.2.1 生成數位變焦影像 36
4.2.2 原方法與本篇方法錯誤率之比較 36
4.3 真實影像立體匹配 46
4.3.1 介面擷取光學變焦影像 46
4.3.2 原方法與本篇方法錯誤率之比較 48
5 結論與未來展望 57
參考文獻 58
Appendices 63
A. Middlebury dataset 立體匹配 64
B. 真實影像立體匹配 73

[1] M. Okutomi and T. Kanade, “A multiple-baseline stereo”, IEEE Trans. Pattern Anal. Mach. Intell., vol. 15, no. 4, pp. 353–363, Apr. 1993.

[2] HongJiang Zhang, Atreyi Kankanhalli, and Stephen W. Smoliar, “Automatic partitioning of full-motion video”, Multimedia Syst., vol. 1, no. 1, pp. 10–28, Jan. 1993.

[3] Szeliski Daniel, Scharstein Richard and Hirschmüller Heiko, “Middlebury”, vision.middlebury.edu/stereo/.

[4] Charles T. Loop and Zhengyou Zhang, “Computing rectifying homographies for stereo vision.”, in CVPR. 1999, pp. 1125–1131, IEEE Computer Society.

[5] Andrea Fusiello, Emanuele Trucco, and Alessandro Verri, “A compact algorithm for rectification of stereo pairs”, Mach. Vision Appl., vol. 12, no. 1, pp. 16–22, July 2000.

[6] Zhengyou Zhang, “Determining the epipolar geometry and its uncertainty: A review”, Int. J. Comput. Vision, vol. 27, no. 2, pp. 161–195, Apr. 1998.

[7] Daniel Scharstein and Richard Szeliski, “A taxonomy and evaluation of dense two-frame stereo correspondence algorithms”, Int. J. Comput. Vision, vol. 47, no. 1-3, pp. 7–42, Apr. 2002.

[8] Aaron F. Bobick and Stephen S. Intille, “Large occlusion stereo.”, International Journal of Computer Vision, vol. 33, no. 3, pp. 181–200, 1999.

[9] Olga Veksler, “Stereo correspondence by dynamic programming on a tree.”, in CVPR (2). 2005, pp. 384–390, IEEE Computer Society.

[10] Qingxiong Yang, Liang Wang, and N. Ahuja, “A constant-space belief propagation algorithm for stereo matching”, in Computer Vision and Pattern Recognition (CVPR), 2010 IEEE Conference on, June 2010, pp. 1458–1465.

[11] Pedro F. Felzenszwalb and Daniel P. Huttenlocher, “Efficient belief propagation for early vision”, Int. J. Comput. Vision, vol. 70, no. 1, pp. 41–54, Oct. 2006.

[12] William T. Freeman, Egon C. Pasztor, and Owen T. Carmichael, “Learning low-level vision”, Int. J. Comput. Vision, vol. 40, no. 1, pp. 25–47, Oct. 2000.

[13] Andreas Klaus, Mario Sormann, and Konrad Karner, “Segment-based stereo matching using belief propagation and a self-adapting dissimilarity measure”, in Proceedings of the 18th International Conference on Pattern Recognition -Volume 03, Washington, DC, USA, 2006, ICPR ’06, pp. 15–18, IEEE Computer Society.

[14] Jian Sun, Nan-Ning Zheng, and Heung-Yeung Shum, “Stereo matching using belief propagation”, IEEE Trans. Pattern Anal. Mach. Intell., vol. 25, no. 7, pp. 787–800, July 2003.

[15] Yi Deng, Qiong Yang, Xueyin Lin, and Xiaoou Tang, “A symmetric patch-based correspondence model for occlusion handling.”, in ICCV. 2005, pp. 1316–1322, IEEE Computer Society.

[16] Yuri Boykov, Olga Veksler, and Ramin Zabih, “Fast approximate energy minimization via graph cuts”, IEEE Trans. Pattern Anal. Mach. Intell., vol. 23, no. 11, pp. 1222–1239, Nov. 2001.

[17] Margrit Bleyer, Michael and Gelautz, “Graph-based surface reconstruction from stereo pairs using image segmentation.”, in SPIE Symposium on Electronic Imaging 2005 (Videometrics VIII), 2005, pp. 288–299.

[18] Li Hong and G. Chen, “Segment-based stereo matching using graph cuts”, in Computer Vision and Pattern Recognition, 2004. CVPR 2004. Proceedings of the 2004 IEEE Computer Society Conference on, June 2004, vol. 1, pp. I–74–I–81 Vol.1.

[19] Heiko Hirschmuller, “Accurate and efficient stereo processing by semi-global matching and mutual information”, in Proceedings of the 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR’05) - Volume 2 - Volume 02, Washington, DC, USA, 2005, CVPR ’05, pp. 807–814, IEEE Computer Society.

[20] Heiko Hirschmuller, “Stereo processing by semiglobal matching and mutual information”, IEEE Trans. Pattern Anal. Mach. Intell., vol. 30, no. 2, pp. 328– 341, Feb. 2008.

[21] Yair Weiss and William T. Freeman, “On the optimality of solutions of the max-product belief propagation algorithm in arbitrary graphs”, 2001.

[22] Gunilla Borgefors, “Distance transformations in digital images”, Comput. Vision Graph. Image Process., vol. 34, no. 3, pp. 344–371, June 1986.

[23] Richard I. Hartley, “In defense of the eight-point algorithm”, IEEE Trans. Pattern Anal. Mach. Intell., vol. 19, no. 6, pp. 580–593, June 1997.

[24] Roger Y. Tsai, “Radiometry”, chapter A Versatile Camera Calibration Technique for High-accuracy 3D Machine Vision Metrology Using Off-the-shelf TV Cameras and Lenses, pp. 221–244. Jones and Bartlett Publishers, Inc., USA, 1992.

[25] Zhengyou Zhang, “A flexible new technique for camera calibration”, IEEE Trans. Pattern Anal. Mach. Intell., vol. 22, no. 11, pp. 1330–1334, Nov. 2000.

[26] J.Y. Bouguet, “Matlab camera calibration toolbox”, 2000.

[27] R. I. Hartley and A. Zisserman, Multiple View Geometry in Computer Vision, Cambridge University Press, ISBN: 0521540518, second edition, 2004.

[28] David G. Lowe, “Distinctive image features from scale-invariant keypoints”, Int. J. Comput. Vision, vol. 60, no. 2, pp. 91–110, Nov. 2004.

[29] Herbert Bay, Tinne Tuytelaars, and Luc Van Gool, “Surf: Speeded up robust features”, in Computer vision–ECCV 2006, pp. 404–417. Springer, 2006.

[30] Luo Juan and Oubong Gwun, “A comparison of sift, pca-sift and surf”, International Journal of Image Processing (IJIP), vol. 3, no. 4, pp. 143–152, 2009.

[31] Yan Ke and Rahul Sukthankar, “Pca-sift: A more distinctive representation for local image descriptors”, in Proceedings of the 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, Washington, DC, USA, 2004, CVPR’04, pp. 506–513, IEEE Computer Society.

[32] Martin A Fischler and Robert 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.

[33] Wei-Tsung Hsiao, Jing-Jang Leou, and Han-Hui Hsiao, “Super-resolution reconstruction for binocular 3d data”, in Pattern Recognition (ICPR), 2014 22nd International Conference on. IEEE, 2014, pp. 4206–4211.