臺灣博碩士論文加值系統

立體匹配一直是計算機視覺的研究焦點，許多機器視覺應用依賴於立體匹配演算法的精確結果。然而，困難的環境條件如光照的差異，嚴重地影響立體匹配演算法的性能。
在本篇論文中，針對非理想光照條件造成立體匹配效果不佳的問題，我們提出了一個在 HSL(色相-飽和度-亮度)顏色空間下融合多特徵的立體匹配計算方法，其結合了亮度梯度資訊、加權距離資訊和色彩非相似性度量。另外，為了提升計算結果的品質，我們同時提出一個改進的最小生成樹代價聚合方法。在實驗中我們採用一系列不同光照條件的圖片集，對本文算法與現有算法進行對比驗證。實驗結果證明，在非理想的光照條件下，我們的方法仍然能夠得到比較理想的視差圖。

Stereo matching has remained in the focus of the computer vision for a few decades. Many machine-vision applications rely on the accurate results of stereo matching algorithms.
However, difficult environmental conditions, such as differentiations in illumination, heavily affect the performance of stereo matching algorithms.
In this paper, we propose a multi-feature matching cost computation in Hue-Saturation-Luminosity (HSL) color space. By combining intensity gradient with weighted distance information and chromatic dissimilarity measure, the proposed method able to tolerate the illumination variation, and also a modified minimum spanning tree cost aggregation is proposed to effectively produce improved disparity maps. Experimental results for a variety of lighting conditions are shown the proposed method which is suitable for non-ideal illumination conditions.

摘要................................................. i
Abstract ............................................ ii
誌謝 ................................................ iii
Content ............................................. iv
List of Figure ...................................... vi
List of Table ....................................... viii
Chapter 1 Introduction .............................. 1
1.1 Motivation and Objective ........................ 3
1.2 Organization of this Thesis ..................... 3
Chapter 2 Background and Related Work ............... 5
2.1 Stereo Matching Flow ............................ 5
2.2 Related Work .................................... 9
2.2.1 Cross-Scale Cost Aggregation................... 9
2.2.2 Tree Structure Cost Aggregation ................14
2.2.3 Normalized Cross Correlation Method ............17
Chapter 3 Proposed Method ............................20
3.1 Overview of the Proposed Method ................. 20
3.2 Image Pre-processing ............................ 22
3.3 Matching Cost Computation with Multi-feature .... 24
3.3.1 Intensity Gradient Information .................24
3.3.2 Distance Information ...........................25
3.3.3 Chromatic Dissimilarity Measure ................27
3.4 Modified Tree Structure Cost Aggregation .........29
3.5 Disparity Refinement ............................ 32
Chapter 4 Experimental Results .......................33
4.1 Evaluation with the Standard Test Sequence .......33
4.2 Performance Measurement ......................... 36
4.2.1 Compare with State-of-the-art Methods ..........36
4.2.2 Compare with Normalized Cross Correlation Methods ......................................................43
4.3 Hardware Implementation Measure ................. 49
Chapter 5 Conclusion and Future Work .................50
Reference.............................................51

[1] Richard Hartley and Andrew Zisserman, “Multiple View Geometry in Computer Vision”, Cambridge University Press, 2000.
[2] N. Hautiere, R. Labayrade, M. Perrollaz, et al., “Road scene analysis by stereovision: a robust and quasi-dense approach.” in Proc IEEE Int. Conf. Control, Automation,
Robotics and Vision, 2006, pp. 1-6.
[3] N. Suganuma, N. Fujiwara, “An obstacle extraction method using virtual disparity image.” in Proc IEEE Int. Conf. Intelligent Vehicles Symposium, 2007, pp. 456-461.
[4] D. Marr, T. Poggio, “Cooperative computation of stereo disparity.” in Science, vol.194, no.4262, pp. 283-287, Oct. 1976.
[5] R. Munoz Salinas, E. Aguirre, M. Garcia Silvente, “People detection and tracking using stereo vision and color.” in Image and Vision Computing, vol.25, no.6, pp. 995-1007, June 2007.
[6] D. Murray, J.J. Little, “Using real-time stereo vision for mobile robot navigation.” in Autonomous Robots, vol.8, no.2, pp. 161-171, April 2000.
[7] R. Sim, J.J. Little, “Autonomous vision-based robotic exploration and mapping using hybrid maps and particle filters.” in Image and Vision Computing, vol.27, no.1-2, pp.
167-177, Jan. 2009.
[8] D. Scharstein and R. Szeliski, “A taxonomy and evaluation of dense two -frame stereo correspondence algorithms.” in International Journal of Computer Vision, vol. 47, no.1-3, pp. 7-42, April-June 2002.
[9] R. Zabih and J. Woodfill, “Non-parametric Local Transforms for Computing Visual Correspondence.” in Proceedings of third European Conference on Computer Vision,
vol. 2, pp. 151–158, 1994.
[10] Y . Boykov, O. Veksler, and R. Zabih, “Fast Approximate Energy Minimization via Graph Cuts.” IEEE transactions on pattern analysis and machine intelligence, pp. 1222-1239,
2001.
[11] P . Fua, “Combining stereo and monocular information to compute dense depth maps that preserve depth discontinuities.” 12th. Int. Joint Conf. on Artificial Intelligence, pp. 1292–1298, 1993.
[12] M . Hariyama, H. Sasaki, and M.Kameyama, “Architecture of a stereo matching VLSI processor based on hierarchically parallel memory access.” The 2004 47th Midwest Symposium on Circuits and Systems, vol 2, pp. II245- II247, 2004.
[13] P . Felzenszwalb and D. Huttenlocher. “Efficient belief propagation for early vision.” In CVPR, 2004.
[14] F . Fleuret and D. Geman. “Coarse-to-fine face detection.” Inter. J. Comp. Vision, 41:85–107, 2001.
[15] D. Marr and T. Poggio. “A computational theory of human stereo vision.” Proceedings of the Royal Society of London. Series B. Biological Sciences, 1979.
[16] G. Van Meerbergen, M. Vergauwen, M. Pollefeys, and L. Van Gool. “A hierarchical symmetric stereo algorithm using dynamic programming.” IJCV, 2002.
[17] Q. Yang, L. Wang, R. Yang, H. Stewenius, and D. Nister. “Stereo matching with color-weighted correlation, hierarchical belief propagation, and occlusion handling.”
TPAMI, 2009.
[18] W . Hu, K. Zhang, L. Sun, and S. Yang. “Comparisons reducing for local stereo matching using hierarchical structure.” In ICME, 2013.
[19] M . Sizintsev. “Hierarchical stereo with thin structures and transparency.” In CCCRV, 2008.
[20] Zhang, K., Fang, Y., Min, D., Sun, L., Yang, S., Yan, S., & Tian, Q. “Cross-scale cost aggregation for stereo matching.” In Proceedings of the IEEE Conference on Computer
Vision and Pattern Recognition (pp. 1590-1597). 2014.
[21] C. Rhemann, A. Hosni, M. Bleyer, C. Rother, and M. Gelautz, “Fast cost-volume filtering for visual correspondence and beyond.” IEEE Conference on Computer Vision and Pattern Recognition, pp. 3017-3028, June 2011.
[22] Tikhonov regularization, https://en.wikipedia.org/wiki/Tikhonov_regularization
[23] Q. Yang. “A non-local cost aggregation method for stereo matching.” IEEE Conference on Computer Vision and Pattern Recognition, 2012:1402-1409.
[24] J. Banks and P. Corke, “Quantitative evaluation of matching methods and validity measures for stereo vision,” The International Journal of Robotics Research, vol. 20, no. 7, pp. 512–532, 2001.
[25] Sun, C., “Fast stereo matching using rectangular subregioning and 3D maximum-surface techniques.” Int. J.Comput. Vis. 47, 2002.
[26] Sun, C., Seleg, S., “Fast panoramic stereo matching using cylindrical maximum surfaces.” IEEE Trans. Syst., 2004.
[27] Di Stefano, Luigi, Stefano Mattoccia. “ZNCC-based template matching using bounded partial correlation.” Pattern recognition letters 26, 2005.
[28] R.C. Gonzalez, R.E. Woods, “Digital Image Processing.” Addison-Wesley Longman Publishing Co., Inc., Boston, MA, USA, 1992. 573607.
[29] K. He, J. Sun, and X. Tang, “Guided image filtering.” in ECCV, 2010.
[30] W. S. Jang and Y. S. Ho, “Discontinuity Preserving Disparity Estimation with Occlusion Handling.” Journal of Visual Communication and Image Representation, vol. 25, no. 7, pp. 1595-1603, Oct. 2014.
[31] Middlebury Stereo Vision Page, http://www.vision.middlebury.edu/stereo/
[32] C. Tomasi and R. Manduchi, “Bilateral filtering for gray and color images,” IEEE Conference on Computer Vision, pp. 839-846, Jan. 1998.
[33] E. Binaghi, I. Gallo, G. Marino, M. Raspanti, “Neural adaptive stereo matching.”Pattern Recognition Letters 25, 1743–1758, 2004.
[34] P. Corke, Machine vision toolbox,
http://petercorke.com/wordpress/toolboxes/machine-vision-toolbox
[35] Xilinx, https://www.xilinx.com/products/boards-and-kits/ek-u1-zcu102-g.html
[36] Tara Akhavan, Hyunjin Yoo, and Margrit Gelautz. “A framework for HDR stereo matching using multi-exposed images.” Proceedings of HDRi2013-First International
Conference and SME Workshop on HDR imaging. 2013.
[37] Tara Akhavan, Hannes Kaufmann. “Backward compatible HDR stereo matching: a hybrid tone-mapping-based framework.” EURASIP Journal on Image and Video Processing, Volume 36, 2015.
[38] D. Scharstein and R. Szeliski. “High-accuracy stereo depth maps using structured light.”IEEE Computer Society Conference on Computer Vision and Pattern Recognition
(CVPR 2003), volume 1, pages 195-202, 2003.
[39] D. Scharstein and C. Pal. “Learning conditional random fields for stereo.” IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR 2007), 2007.
[40] H. Hirschmüller and D. Scharstein. “Evaluation of cost functions for stereo matching.”IEEE Computer Society Conference on Computer Vision and Pattern Recognition
(CVPR 2007), 2007.
[41] Hirschmuller, Heiko, and Daniel Scharstein. “Evaluation of stereo matching costs on images with radiometric differences.” IEEE transactions on pattern analysis and machine intelligence 31.9: 1582-1599, 2009.
[42] Patil, S., Nadar, J. S., Gada, J., Motghare, S., & Nair, S. S. “Comparison of various stereo vision cost aggregation methods.” International Journal of Engineering and Innovative Technology, 2(8), 222-226, 2013.
[43] M. Bleyer, S. Chambon. “Does Color Really Help in Dense Stereo Matching?” Talk: International Symposium 3D Data Processing, Visualization and Transmission (3DPVT),
2010.
[44] Brockers, Roland. “Cooperative stereo matching with color-based adaptive local support.” Computer Analysis of Images and Patterns. Springer Berlin/Heidelberg, 2009.
[45] Galar, M., Jurio, A., Lopez-Molina, C., Paternain, D., Sanz, J., & Bustince, H.
“Aggregation functions to combine RGB color channels in stereo matching.” Optics express, 21(1), 1247-1257, 2013.
[46] Konolige, Kurt. “Small vision systems: Hardware and implementation.” Robotics research. Springer, London, 203-212, 1998.
[47] Mei, X., Sun, X., Zhou, M., Jiao, S., Wang, H., & Zhang, X. “On building an accurate stereo matching system on graphics hardware.” In Computer Vision Workshops (ICCV
Workshops), 2011.
[48] Bae, Kyeong-ryeol, and Byungin Moon. “An accurate and cost-effective stereo matching algorithm and processor for real-time embedded multimedia systems.” Multimedia Tools
and Applications, 2016.