立體視覺 (Stereo Vision) 是以模擬人類雙眼所開發出的一套深度視覺預測演算法，藉由兩台攝影機拍攝場景，找尋兩張影像中匹配像素，進一步產生影像之視差圖。目前立體視覺之演算法主要分為三種類型：區域性、全域性及融合兩種特徵的半全域性演算法，區域性較容易以硬體架構設計，但是準確率會相對較低，可視性也較差。相反地，全域性及半全域性之演算法較難以硬體架構及流程開發，但是準確率能夠高出許多。

在硬體架構開發上，主要有四個需要考量的地方：準確率、硬體資源、運算速度及功耗，本文藉由觀察影像及深度之關係，開發出一套利於硬體架構開發的視差修補演算法，利用邊緣偵測技術，針對同一個物體或背景內之視差值進行可信任視差統計，再對邊緣間利用最大數量之可信任視差對不可信任視差進行修補。本文開發出一套視差預測之硬體架構，導入引導影像濾波器 (Guided Image Filter) 技術進行代價優化，加入代價計算條件以考慮光照不均勻問題，並且套用本文開發之視差修補演算法，在 Xilinx FPGA 開發版上進行模擬，使用Middlebury V2、V3等測試圖進行準確率測試，不僅能得到較好的準確率，也能得到較低硬體資源使用率。

Stereo Vision (Stereo Vision) is a depth prediction algorithm that uses two cameras to capture the same scene and find matching pixels in two images to further generate a disparity map of the image. At present, the algorithms for stereo vision are mainly divided into three types: local, global, and semi-global method. Local stereo matching algorithms are easier to design with hardware architecture, but the accuracy will be relatively low, and visibility also poor. Conversely, global and semi-global methods are more difficult to develop in hardware architecture and processes, but the accuracy can be much higher.

In the development of hardware architecture, there are four key points to be considered: accuracy, hardware resources, computing speed and power consumption. By observing the relationship between image and depth, we introduce a disparity refinement methodology that is conducive to the development of hardware architecture. This algorithm builds up a histogram of valid disparity by using edge detection technology, and then repairs the invalid disparities by the valid disparity which has the largest number between two edge pixels. This paper presents a low-cost and high-quality hardware design which introduces Guided Image Filter technology for matching cost optimization, adds cost calculation conditions to solve uneven lighting problem, and applies the proposed disparity refinement algorithm. It is implemented on Xilinx FPGA devices, and use Middlebury V2, V3 to evaluate the accuracy. Compared with other state-of-the-art designs, our method can not only get a better accuracy rate, but also get a lower hardware resource utilization rate.

D. Scharstein, R. Szeliski and R. Zabih, "A taxonomy and evaluation of dense two-frame stereo correspondence algorithms," Proceedings IEEE Workshop on Stereo and Multi-Baseline Vision (SMBV 2001), Kauai, HI, USA, 2001, pp. 131-140, doi: 10.1109/SMBV.2001.988771.

Kok, Kai Yit & Rajendran, Parvathy. (2019). A Review on Stereo Vision Algorithm: Challenges and Solutions. 13. 10.37936/ecti-cit.2019132.194324.

Kuk-Jin Yoon and In So Kweon, "Adaptive support-weight approach for correspondence search," in IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 28, no. 4, pp. 650-656, April 2006, doi: 10.1109/TPAMI.2006.70.

N. Y. Chang, T. Tsai, B. Hsu, Y. Chen and T. Chang, "Algorithm and Architecture of Disparity Estimation With Mini-Census Adaptive Support Weight," in IEEE Transactions on Circuits and Systems for Video Technology, vol. 20, no. 6, pp. 792-805, June 2010, doi: 10.1109/TCSVT.2010.2045814.

C. Ttofis and T. Theocharides, "Towards accurate hardware stereo correspondence: A real-time FPGA implementation of a segmentation-based adaptive support weight algorithm," 2012 Design, Automation & Test in Europe Conference & Exhibition (DATE), Dresden, 2012, pp. 703-708, doi: 10.1109/DATE.2012.6176561.

J. Kowalczuk, E. T. Psota and L. C. Perez, "Real-Time Stereo Matching on CUDA Using an Iterative Refinement Method for Adaptive Support-Weight Correspondences," in IEEE Transactions on Circuits and Systems for Video Technology, vol. 23, no. 1, pp. 94-104, Jan. 2013, doi: 10.1109/TCSVT.2012.2203200.

D. Chen, M. Ardabilian and L. Chen, "A Fast Trilateral Filter-Based Adaptive Support Weight Method for Stereo Matching," in IEEE Transactions on Circuits and Systems for Video Technology, vol. 25, no. 5, pp. 730-743, May 2015, doi: 10.1109/TCSVT.2014.2361422.

C. Banz, S. Hesselbarth, H. Flatt, H. Blume and P. Pirsch, "Real-time stereo vision system using semi-global matching disparity estimation: Architecture and FPGA-implementation," 2010 International Conference on Embedded Computer Systems: Architectures, Modeling and Simulation, Samos, 2010, pp. 93-101, doi: 10.1109/ICSAMOS.2010.5642077.

W. Wang, J. Yan, N. Xu, Y. Wang and F. Hsu, "Real-time high-quality stereo vision system in FPGA," 2013 International Conference on Field-Programmable Technology (FPT), Kyoto, 2013, pp. 358-361, doi: 10.1109/FPT.2013.6718387.

Y. Li, C. Yang, W. Zhong, Z. Li and S. Chen, "High throughput hardware architecture for accurate semi-global matching," 2017 22nd Asia and South Pacific Design Automation Conference (ASP-DAC), Chiba, 2017, pp. 641-646, doi: 10.1109/ASPDAC.2017.7858396.

L. F. S. Cambuim, J. P. F. Barbosa and E. N. S. Barros, "Hardware module for low-resource and real-time stereo vision engine using semi-global matching approach," 2017 30th Symposium on Integrated Circuits and Systems Design (SBCCI), Fortaleza, 2017, pp. 53-58.

J. Wang, Z. Li, L. Yao, S. Chen and F. Wu, "Low-Resource Hardware Architecture for Semi-Global Stereo Matching," 2019 IEEE International Symposium on Circuits and Systems (ISCAS), Sapporo, Japan, 2019, pp. 1-4, doi: 10.1109/ISCAS.2019.8702234.

A. Hosni, M. Bleyer, C. Rhemann, M. Gelautz and C. Rother, "REal-time local stereo matching using guided image filtering," 2011 IEEE International Conference on Multimedia and Expo, Barcelona, 2011, pp. 1-6, doi: 10.1109/ICME.2011.6012131.

S. Hsiao and C. Chang, "Hardware design of disparity computation for stereo vision using guided image filtering," 2018 International Symposium on VLSI Design, Automation and Test (VLSI-DAT), Hsinchu, 2018, pp. 1-4, doi: 10.1109/VLSI-DAT.2018.8373266.

C. Ttofis, C. Kyrkou and T. Theocharides, "A Low-Cost Real-Time Embedded Stereo Vision System for Accurate Disparity Estimation Based on Guided Image Filtering," in IEEE Transactions on Computers, vol. 65, no. 9, pp. 2678-2693, 1 Sept. 2016, doi: 10.1109/TC.2015.2506567.

G. S. Hong, B. G. Kim, "A local stereo matching algorithm based on weighted guided image filtering for improving the generation of depth range images, " in Displays 2017, vol. 49, pp. 80–87. 13 July 2017.

C. K. Vala et al., "High-Speed Low-Complexity Guided Image Filtering-Based Disparity Estimation," in IEEE Transactions on Circuits and Systems I: Regular Papers, vol. 65, no. 2, pp. 606-617, Feb. 2018, doi: 10.1109/TCSI.2017.2729084.

C. Ttofis and T. Theocharides, "High-quality real-time hardware stereo matching based on guided image filtering," 2014 Design, Automation & Test in Europe Conference & Exhibition (DATE), Dresden, 2014, pp. 1-6, doi: 10.7873/DATE.2014.369.

Y. Li, L. Claesen, K. Huang and M. Zhao, "A Real-Time High-Quality Complete System for Depth Image-Based Rendering on FPGA," in IEEE Transactions on Circuits and Systems for Video Technology, vol. 29, no. 4, pp. 1179-1193, April 2019, doi: 10.1109/TCSVT.2018.2825022.

M. Dehnavi, M. Eshghi, "Cost and power efficient FPGA based
stereo vision system using directional graph transform, " in Journal of
Visual Communication and Image Representation, Volume 56,
October 2018, Pages 106-115.

K. Zhang, J. Lu and G. Lafruit, "Cross-Based Local Stereo Matching Using Orthogonal Integral Images," in IEEE Transactions on Circuits and Systems for Video Technology, vol. 19, no. 7, pp. 1073-1079, July 2009, doi: 10.1109/TCSVT.2009.2020478.

Shan, Yi & Hao, Yuchen & Wang, Wenqiang & Wang, Yu & Chen, Xu & Yang, Huazhong & Luk, Wayne. (2014). Hardware Acceleration for an Accurate Stereo Vision System Using Mini-Census Adaptive Support Region. ACM Transactions on Embedded Computing Systems. 13. pp. 1-24. 10.1145/2584659.

Zitnick, C. & Kang, Sing Bing & Uyttendaele, Matt & Winder, Simon & Szeliski, Richard. (2004). High-Quality Video View Interpolation Using a Layered Representation. ACM Trans. Graph.. 23. 600-608. 10.1145/1015706.1015766.

A. Klaus, M. Sormann and K. Karner, "Segment-Based Stereo Matching Using Belief Propagation and a Self-Adapting Dissimilarity Measure," 18th International Conference on Pattern Recognition (ICPR'06), Hong Kong, 2006, pp. 15-18, doi: 10.1109/ICPR.2006.1033.

R. Brockers, M. Hund, and B. Mertsching, “Stereo vision using
cost-relaxation with 3D support regions,” in Image and Vision
Computing New Zealand, 2005.

Pons, J., Keriven, R. & Faugeras, O. Multi-View Stereo Reconstruction and Scene Flow Estimation with a Global Image-Based Matching Score. Int J Comput Vision 72, 179–193 (2007).

H. Xu, X. Chen, H. Liang, S. Ren, Y. Wang and H. Cai, "CrossPatch-Based Rolling Label Expansion for Dense Stereo Matching," in IEEE Access, vol. 8, pp. 63470-63481, 2020, doi: 10.1109/ACCESS.2020.2985106.

X. Ye, J. Li, H. Wang, H. Huang and X. Zhang, "Efficient Stereo Matching Leveraging Deep Local and Context Information," in IEEE Access, vol. 5, pp. 18745-18755, 2017, doi: 10.1109/ACCESS.2017.2754318.

H. Park and K. M. Lee, "Look Wider to Match Image Patches With Convolutional Neural Networks," in IEEE Signal Processing Letters, vol. 24, no. 12, pp. 1788-1792, Dec. 2017, doi: 10.1109/LSP.2016.2637355.

H. Park and K. M. Lee, "Look Wider to Match Image Patches With Convolutional Neural Networks," in IEEE Signal Processing Letters, vol. 24, no. 12, pp. 1788-1792, Dec. 2017, doi: 10.1109/LSP.2016.2637355.

K. He, J. Sun and X. Tang, "Guided Image Filtering," in IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 35, no. 6, pp. 1397-1409, June 2013, doi: 10.1109/TPAMI.2012.213.

M. Gong and Y. Yang, "Real-Time Stereo Matching Using Orthogonal Reliability-Based Dynamic Programming," in IEEE Transactions on Image Processing, vol. 16, no. 3, pp. 879-884, March 2007, doi: 10.1109/TIP.2006.891344.

Vladimir Kolmogorov, Pascal Monasse, Pauline Tan. Kolmogorov and Zabih's Graph Cuts Stereo Matching Algorithm. Image Processing On Line, IPOL - Image Processing on Line, 2014, 4, pp.220-251. ⟨10.5201/ipol.2014.97⟩. ⟨hal-01074878⟩

Stereo - The Middlebury Computer Vision Pages. [Online]. 2014, https://vision.middlebury.edu/stereo/eval3/

Jin, Minxi & Maruyama, Tsutomu. (2014). Fast and Accurate Stereo Vision System on FPGA. ACM Transactions on Reconfigurable Technology and Systems (TRETS). 7. 10.1145/2567659.