研究生: |
黃正瑜 Zheng-Yu Huang |
---|---|
論文名稱: |
基於智慧手錶的即時羽球球種辨識 Badminton Stroke Recognition Based on A Smartwatch |
指導教授: |
林淵翔
Yuan-Hsiang Lin |
口試委員: |
陳儷今
Li-Chin Chen 阮聖彰 Shanq-Jang Ruan 陳維美 Wei-Mei Chen |
學位類別: |
碩士 Master |
系所名稱: |
電資學院 - 電子工程系 Department of Electronic and Computer Engineering |
論文出版年: | 2021 |
畢業學年度: | 109 |
語文別: | 中文 |
論文頁數: | 67 |
中文關鍵詞: | 羽球 、慣性感測器 、擊球偵測 、球種辨識 、機器學習 |
外文關鍵詞: | badminton, inertial sensor, stroke detection, stroke types recognition, machine learning |
相關次數: | 點閱:452 下載:0 |
分享至: |
查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報 |
隨著運動科學的興起,運動數據的量化分析對運動選手的訓練逐漸重要。在羽球運動項目中,傳統上主要透過教練的監督與錄製影像的方式進行選手的訓練評估,但此方法因錄影設備較為昂貴以及架設困難,且容易受到拍攝角度與範圍的影響,無法有效分辨出各種球路動作。
使用慣性感測器紀錄球路,除了能避免攝影機架設死角問題,也因其重量輕巧、攜帶方便,使之成為羽球動作分析中新興的感測方式,加上近年來配戴穿戴式裝置從事運動已成為一股潮流。因此,我們建立了一套基於智慧手錶的羽球球種辨識系統,利用智慧手錶內建的加速度計與陀螺儀擷取球員的擊球動作後,使用機器學習進行數據分析分類,即時記錄使用者的球種資訊和擊球次數。本論文結合滑動視窗法和SVM機器學習演算法來即時偵測擊球動作以及辨識球種。使用6位受測者的資料來建立分類8種球種和分類11種球種的共通模型,並透過3位受測者做即時系統的驗證。即時驗證結果顯示擊球偵測的辨識準確度為100%,分類8種球種和分類11種球種的共通模型辨識平均準確度分別為98%與95%。與個人模型相比,分類8種球種與分類11種球種的個人模型辨識平均準確度分別為97%與96%。本論文所提出的系統能將擊球動作數據化,讓使用者即時知道自己的運動狀態,輔助使用者進行訓練。
With the rise of sports science, quantitative analysis of sports data has become increasingly important for athletes' training. Traditionally in badminton sports, the evaluation of training is mainly carried out through the supervision of coaches and recording videos. However, this method cannot distinguish the variety of stroke types efficiently in real-time due to the expensive camera and the set-up difficulty, and it is easily affected by the shooting angle and range.
The use of motion devices to record stroke movements can avoid blind spots and is conveniently portable because it is lightweight, making it an emerging sensing method in badminton motion analysis. Involving wearable devices for sports in recent years has become a trend. Therefore, we established a badminton stroke recognition system to capture the stroke movements using the built-in accelerometer and gyroscope on a smartwatch to classify the stroke data with the machine learning method. Apart from classification, our system can record the player's stroke type information and the number of shots in real-time.
Our research method combines the sliding window method and SVM machine learning algorithm to detect the player's stroke and recognize the badminton stroke types in real-time. To train the general model for 8 and 11 stroke types, we used six subjects as training data and three subjects for real-time verification. The results show that the accuracy of the stroke detection is 100%, the average accuracy of the general model of 8 stroke types is 98%, and the average accuracy of 11 stroke types is 95%. Compared with the personal model, the average accuracy of the personal model of 8 stroke types is 97%, and the average accuracy of 11 stroke types is 96%. The system proposed in this thesis can digitize the badminton stroke to know their status in real-time and assist users in training.
[1] M. Phomsoupha and G. Laffaye, "The Science of Badminton: Game Characteristics, Anthropometry, Physiology, Visual Fitness and Biomechanics," Sports Medicine, vol. 45, pp. 473-495, Apr. 2015, doi: 10.1007/s40279-014-0287-2.
[2] J.-Y. Guillain, Badminton : An Illustrated History (French Edition) Publibook (in French), 2004.
[3] 教育部體育署, "中華民國109年運動現況調查結案報告書."
[4] M. Abou Elmagd, "Benefits, Need and Importance of Daily Exercise," International Journal of Physical Education, Sports and Health, vol. 3, no. 5, pp. 22-27, 2016.
[5] D. Lo and K. Stark, "Sports Performance Series: The Badminton Overhead Shot," National Strength & Conditioning Association Journal, vol. 13, no. 4, p. 6, 1991, doi: 10.1519/0744-0049(1991)013<0006:TBOS>2.3.CO;2.
[6] K. Teu, W. Kim, J. Tan, and F. Fuss, "Using Dual Euler Angles for the Analysis of Arm Movement During the Badminton Smash," Sports Engineering, vol. 8, pp. 171-178, Jan. 2005, doi: 10.1007/BF02844017.
[7] D. Cabello Manrique and J. J. González-Badillo, "Analysis of the Characteristics of Competitive Badminton," British Journal of Sports Medicine, vol. 37, no. 1, pp. 62-6, Feb. 2003, doi: 10.1136/bjsm.37.1.62.
[8] O. D. Lara and M. A. Labrador, "A Survey on Human Activity Recognition using Wearable Sensors," IEEE Communications Surveys & Tutorials, vol. 15, no. 3, pp. 1192-1209, Nov. 2013, doi: 10.1109/SURV.2012.110112.00192.
[9] D. B. Waddell and B. A. Gowitzke, "Biomechanical Principles Applied to Badminton Power Strokes," 2000.
[10] W. R. Thompson, "Worldwide Survey of Fitness Trends for 2020," ACSM'S Health & Fitness Journal, vol. 23, no. 6, pp. 10-18, Nov. 2019, doi: 10.1249/fit.0000000000000526.
[11] G. Zhu, C. Xu, and W. Gao, "Action Recognition in Broadcast Tennis Video Using Optical Flow and Support Vector Machine," presented at the Computer Vision in Human-Computer Interaction, Berlin, Heidelberg, 2006.
[12] H. Y. Ting, K. S. Sim, and F. Abas, "Kinect-based Badminton Movement Recognition and Analysis System," International Journal of Computer Science in Sport, vol. 14, pp. 25-41, Jan. 2015.
[13] P. Martin, J. Benois-Pineau, and R. Péteri, "Fine-Grained Action Detection and Classification in Table Tennis with Siamese Spatio-Temporal Convolutional Neural Network," in 2019 IEEE International Conference on Image Processing (ICIP), Sept. 2019, pp. 3027-3028, doi: 10.1109/ICIP.2019.8803382.
[14] Y. Tian, X. Wang, L. Chen, and Z. Liu, "Wearable Sensor-Based Human Activity Recognition via Two-Layer Diversity-Enhanced Multiclassifier Recognition Method," Sensors, vol. 19, no. 9, p. 2039, Apr. 2019, doi: 10.3390/s19092039.
[15] Z. Chen et al., "3D Vision Based Fast Badminton Localization with Prediction and Error Elimination for Badminton Robot," in 2016 12th World Congress on Intelligent Control and Automation (WCICA) Jun. 2016, pp. 3050-3055, doi: 10.1109/WCICA.2016.7578819.
[16] Y. Wang et al., "IoT for Next-Generation Racket Sports Training," IEEE Internet of Things Journal, vol. 5, no. 6, pp. 4558-4566, Dec. 2018, doi: 10.1109/JIOT.2018.2837347.
[17] F. Attal et al., "Physical Human Activity Recognition Using Wearable Sensors," Sensors, vol. 15, pp. 31314-31338, Dec. 2015, doi: 10.3390/s151229858.
[18] Z. Wang, M. Guo, and C. Zhao, "Badminton Stroke Recognition Based on Body Sensor Networks," IEEE Transactions on Human-Machine Systems, vol. 46, no. 5, pp. 769-775, 2016, doi: 10.1109/THMS.2016.2571265.
[19] R. Liu et al., "Table Tennis Stroke Recognition Based on Body Sensor Network," in Internet and Distributed Computing Systems, Cham, Nov. 2019: Springer International Publishing, pp. 1-10, doi: 10.1007/978-3-030-34914-1_1.
[20] Y. Hsu, H. Chang, and Y. Chiu, "Wearable Sport Activity Classification Based on Deep Convolutional Neural Network," IEEE Access, vol. 7, pp. 170199-170212, 2019, doi: 10.1109/ACCESS.2019.2955545.
[21] S. Stancin and S. Tomazic, "Early Improper Motion Detection in Golf Swings Using Wearable Motion Sensors: The First Approach," Sensors, vol. 13, no. 6, pp. 7505-21, Jun. 2013, doi: 10.3390/s130607505.
[22] Y. Wang, Y. Zhao, R. H. M. Chan, and W. J. Li, "Volleyball Skill Assessment Using a Single Wearable Micro Inertial Measurement Unit at Wrist," IEEE Access, vol. 6, pp. 13758-13765, Jan. 2018, doi: 10.1109/ACCESS.2018.2792220.
[23] X. Zhang et al., "Wrist MEMS Sensor for Movements Recognition in Ball Games," in 2019 IEEE 9th Annual International Conference on CYBER Technology in Automation, Control, and Intelligent Systems (CYBER), 2019, pp. 1663-1667, doi: 10.1109/CYBER46603.2019.9066506.
[24] K. Dokic, T. Mesic, and M. Martinović, "Table Tennis Forehand and Backhand Stroke Recognition Based on Neural Network," in Advances in Computing and Data Sciences, 2020, pp. 24-35.
[25] K. Xia et al., "Racquet Sports Recognition Using a Hybrid Clustering Model Learned from Integrated Wearable Sensor," Sensors, vol. 20, p. 1638, Mar. 2020, doi: 10.3390/s20061638.
[26] W. R. Hamilton, "II. On Quaternions; or on a New System of Imaginaries in Algebra," The London, Edinburgh, and Dublin Philosophical Magazine and Journal of Science, vol. 25, no. 163, pp. 10-13, Jul. 1844, doi: 10.1080/14786444408644923.
[27] C. Cortes and V. Vapnik, "Support-vector networks," Machine Learning, vol. 20, no. 3, pp. 273-297, Sept. 1995, doi: 10.1007/BF00994018.
[28] VICTOR勝利體育. "羽球教室." https://www.victorsport.com.tw/badmintonaz.html?cid=139&order=latest (accessed 2014).
[29] Apple Inc. "Apple Watch Series 5 - 技術規格." https://support.apple.com/kb/SP808?viewlocale=zh_TW&locale=zh_TW (accessed Mar. 23, 2021).
[30] T. Steels et al., "Badminton Activity Recognition Using Accelerometer Data," Sensors, vol. 20, no. 17, p. 4685, Aug. 2020, doi: 10.3390/s20174685.
[31] M. Peker et al., "A Novel Hybrid Method for Determining the Depth of Anesthesia Level: Combining ReliefF Feature Selection and Random Forest Algorithm (ReliefF+RF)," in 2015 International Symposium on Innovations in Intelligent SysTems and Applications Sept. 2015, pp. 1-8, doi: 10.1109/INISTA.2015.7276737.
[32] F. Pedregosa et al., "Scikit-learn: Machine Learning in Python," J. Mach. Learn. Res., vol. 12, pp. 2825–2830, 2011.
[33] Scikit-learn. "Scikit-learn Algorithm Cheat-sheet." https://scikit-learn.org/stable/tutorial/machine_learning_map/index.html (accessed 2021).
[34] 羽球筆記. " National Badminton Rating Program分級規則表. "http://www.badmintonnote.com.tw/Rank/Select(accessed 2021)
[35] Torres-Luque Gema, Fernández-García Ángel Iván, Blanca-Torres Juan Carlos, Kondric Miran, and Cabello-Manrique David, "Statistical Differences in Set Analysis in Badminton at the RIO 2016 Olympic Games ," Frontiers in Psychology, vol. 10, p. 731, Apr. 2019, doi: 10.3389/fpsyg.2019.00731.