臺灣博碩士論文加值系統

遠程光電容積脈波（rPPG）技術是一種無接觸、非侵入式的方法，用於提取人體的生理訊號，如心率和血氧濃度等。由於 rPPG 需要處理高分辨率影像並從中提取微小的生理變化訊號，傳統的軟體方法在計算效率和即時性方面面臨挑戰。在本文中，提出一種基於嵌入式可程式邏輯架構（例如 AMD kria kv260）的硬體加速平台，用於優化 rPPG 演算法的執行效率。

本論文設計並實現一個專用硬體加速器，涵蓋數據訊號預處理、平面正交於皮膚（POS）演算法以及後續的訊號過濾處理。與傳統軟體實現相比，這種硬體加速方案在計算效率、即時性和功耗方面具有顯著優勢。實驗結果顯示，使用HRPA硬體加速器，rPPG訊號的處理速度提高13倍，並且在保持高準確度的同時，顯著降低處理延遲。

Remote photoplethysmography (rPPG) technology is a contactless, non-invasive method used to extract physiological signals such as heart rate and blood oxygen concentration from the human body. Due to the need to process high-resolution images and extract subtle physiological signal changes, traditional software-based methods face challenges in terms of computational efficiency and real-time performance. In this paper, we propose a hardware acceleration platform based on an embedded programmable logic architecture (e.g., AMD kria kv260) to optimize the execution efficiency of rPPG algorithms.

We designed and implemented a dedicated hardware accelerator that encompasses data signal preprocessing, the rPPG algorithm(plane orthogonal to skin (POS) algorithm), and subsequent signal filtering processing. Compared to traditional software implementations, this hardware acceleration scheme offers significant advantages in terms of computational efficiency, real-time performance, and power consumption. Experimental results show that with HRPA (Hardware rPPG POS Accelerator), the processing speed of rPPG signals increased by 13 times, while maintaining high accuracy and significantly reducing processing latency.

摘要......i
Abstract......ii
誌謝......iii
目錄......iv
表目錄......vi
圖目錄......vii
第一章 緒論......1
1.1 研究動機......1
1.2 研究背景......3
1.3 研究方法......4
1.4 論文架構......5
第二章 rPPG......6
2.1 rPPG基本原理......6
2.1.1 人臉捕捉......7
2.1.2 ROI的處理與訊號預處理......7
2.2 rPPG演算法......8
2.2.1 Blind source separation (BSS)......8
2.2.2 CHROM......12
2.2.3 PBV......13
2.2.4 2SR......14
2.2.5 POS......16
2.3 rPPG濾波器設計......18
第三章 硬體架構設計與實現......24
3.1 高層次綜合與PYNQ的設計流程......24
3.2 硬體架構......26
3.3 硬體設計......28
3.4 HLS優化方法......31
3.4.1 程式碼優化......31
3.4.2 相關循環指令優化......32
第四章 實驗結果......39
4.1 函數的優化比較......40
4.2 硬體配置後的模擬結果......42
第五章 結論與未來展望......45
參考文獻......46
Extended Abstract......49

[1]World Heath Organization, “Cardiovasculardiseases(CVDs)” Jun, 11, 2021[online], Available:https://www.who.int/news-room/fact-sheets/detail/cardiovascular-diseases-(cvds), Accessed on Jun, 8, 2024.
[2]ministry of health and welfare, “ 112年我國十大死因” Jun, 17, 2024[online], Available: https://www.mohw.gov.tw/cp-16-79055-1.html, Accessed on Jun, 24, 2024.
[3]TrendForce, “預估2028年在穿戴式裝置應用市場可達4.22億美元” Jan, 2, 2024[online], Available:https://www.trendforce.com.tw/presscenter/news/20240122-12007.html.
[4]Verkruysse W, Svaasand LO, Nelson JS, “Remote plethysmographic imaging using ambient light,” Opt Express. 2008 Dec 22;16(26):21434-45. doi: 10.1364/oe.16.021434. PMID: 19104573; PMCID: PMC2717852.
[5]ZCON Telehealth, “FACE AI 看鏡頭，就可知道你的健康狀況! 無接觸，安全快速!”[online], Available: https://www.zconai.com/web/project/.
[6]Nasir Hayat, “Robust Heart rate estimation from facial videos” May, 19, 2019[online], Availible: https://github.com/nasir6/rPPG.
[7]G. Boccignone, D. Conte, V. Cuculo, A. D’Amelio, G. Grossi and R. Lanzarotti, “An Open Framework for Remote-PPG Methods and Their Assessment,” in IEEE Access, vol. 8, pp. 216083-216103, 2020, doi: 10.1109/ACCESS.2020.3040936.
[8]P. Hansen, M. G. Lozano, F. Kamrani and J. Brynielsson, “Real-Time Estimation of Heart Rate in Situations Characterized by Dynamic Illumination using Remote Photoplethy-smography,” 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), Vancouver, BC, Canada, 2023, pp. 6094-6103, doi: 10.1109/CVPRW59228.2023.00649.
[9]M. W. Numan, B. J. Phillips, G. S. Puddy and K. Falkner, “Towards Automatic High-Level Code Deployment on Reconfigurable Platforms: A Survey of High-Level Synthesis Tools and Toolchains,” in IEEE Access, vol. 8, pp. 174692-174722, 2020, doi: 10.1109/ACCESS.2020.3024098.
[10]S. Kolkur, D. Kalbande, P. Shimpi, C. Bapat and J. Jatakia, “Human skin detection using RGB HSV and YCbCr color models”, arXiv:1708.02694, 2017, [online] Available: http://arxiv.org/abs/1708.02694.
[11]V. Kazemi and J. Sullivan, "One millisecond face alignment with an ensemble of regression trees", Proc. IEEE Conf. Comput. Vis. Pattern Recognit., pp. 1867-1874, Jun. 2014.
[12]K. Zhang, Z. Zhang, Z. Li and Y. Qiao, "Joint face detection and alignment using multitask cascaded convolutional networks", IEEE Signal Process. Lett., vol. 23, no. 10, pp. 1499-1503, Oct. 2016.
[13]R. J. Qian, M. I. Sezan and K. E. Matthews, "A robust real-time face tracking algorithm", Proc. Int. Conf. Image Process. (ICIP), vol. 1, pp. 131-135, 1998.
[14]S. Kolkur, D. Kalbande, P. Shimpi, C. Bapat and J. Jatakia, "Human skin detection using RGB HSV and YCbCr color models", arXiv:1708.02694, 2017, [online] Available: http://arxiv.org/abs/1708.02694.
[15]M. Lewandowska, J. Rumiński, T. Kocejko and J. Nowak, “Measuring pulse rate with a Webcam—Non-contact method for evaluating cardiac activity”, Proc. Federated Conf. Comput. Sci. Inf. Syst., pp. 405-410, 2011.
[16]A. Hyvärinen and E. Oja, "Independent component analysis: Algorithms and applications", Neural Netw., vol. 13, no. 4, pp. 411-430, Jun. 2000.
[17]M.-Z. Poh, D. J. McDuff and R. W. Picard, “Non-contact automated cardiac pulse measurements using video imaging and blind source separation”, Opt. Exp., vol. 18, no. 10, pp. 10762-10774, May 2010.
[18]J.-F. Cardoso and A. Souloumiac, "Blind beamforming for non-Gaussian signals", IEE Proc.-F Radar Signal Process., vol. 140, no. 6, pp. 362-370, Dec. 1993.
[19]A. Hyvärinen and E. Oja, "A fast fixed-point algorithm for independent component analysis", Neural Comput., vol. 9, no. 7, pp. 1483-1492, Oct. 1997.
[20]G. De Haan and V. Jeanne, “Robust pulse rate from chrominance-based rPPG”, IEEE Trans. Biomed. Eng., vol. 60, no. 10, pp. 2878-2886, Oct. 2013.
[21]G. de Haan and A. van Leest, “Improved motion robustness of remote-PPG by using the blood volume pulse signature”, Physiol. Meas., vol. 35, no. 9, pp. 1913-1922, Oct. 2014.
[22]W. Wang et al., “A novel algorithm for remote photoplethysmography: Spatial subspace rotation”, IEEE Trans. Biomed. Eng., vol. 63, no. 9, pp. 1974-1984, Sep. 2016.
[23]W. Wang, A. C. den Brinker, S. Stuijk and G. de Haan, “Algorithmic principles of remote PPG”, IEEE Trans. Biomed. Eng., vol. 64, no. 7, pp. 1479-1491, Jul. 2017.
[24]Scipy, “scipy.signal”,Availible: https://github.com/scipy/scipy.
[25]J. K. Kim and J. M. Ahn, "Design of an optimal digital IIR filter for heart rate variability by photoplethysmogram", Int. J. Eng. Res. Technol., vol. 11, no. 12, pp. 2009-2021, 2018.
[26]R. S. Molina, V. Gil-Costa, M. L. Crespo and G. Ramponi, "High-Level Synthesis Hardware Design for FPGA-Based Accelerators: Models, Methodologies, and Frameworks," in IEEE Access, vol. 10, pp. 90429-90455, 2022, doi: 10.1109/ACCESS.2022.3201107.
[27]Simon Urbanek and Aaron Puchert, “OpenCL: Interface allowing R to use OpenCL”, Aug, 8, 2011[online], Availible: http://www.rforge.net/OpenCL/.
[28]J. E. Stone, D. Gohara and G. Shi, "OpenCL: A Parallel Programming Standard for Heterogeneous Computing Systems," in Computing in Science & Engineering, vol. 12, no. 3, pp. 66-73, May-June 2010, doi: 10.1109/MCSE.2010.69.
[29]J. de Fine Licht, M. Besta, S. Meierhans and T. Hoefler, "Transformations of High-Level Synthesis Codes for High-Performance Computing," in IEEE Transactions on Parallel and Distributed Systems, vol. 32, no. 5, pp. 1014-1029, 1 May 2021, doi: 10.1109/TPDS.2020.3039409.
[30]S. Derrien, T. Marty, S. Rokicki and T. Yuki, "Toward Speculative Loop Pipelining for High-Level Synthesis," in IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, vol. 39, no. 11, pp. 4229-4239, Nov. 2020, doi: 10.1109/TCAD.2020.3012866.
[31]A. Morvan, S. Derrien and P. Quinton, "Polyhedral Bubble Insertion: A Method to Improve Nested Loop Pipelining for High-Level Synthesis," in IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, vol. 32, no. 3, pp. 339-352, March 2013, doi: 10.1109/TCAD.2012.2228270.
[32]AMD. “KV260 Starter Kit Power Specifications”, 2024 [online], Availible: https://docs.amd.com/r/en-US/ds986-kv260-starter-kit