重訓可以增進整體的身體健康，讓到一個人有更好的體態，已經提高運動表現。在一個訓練環節內影響重訓的效率的因素主要有4個，分別是運動的重量、次數與組數、進行動作的速度以及使用的負重。前人的研究有使用穿戴式的感測器去偵測運動的種類、做的次數與組數、以及進行動作時的速度。可是，想要去偵測重訓時使用的負重，需要特定的運動設備或負重。我們的研究MuscleSense，是一個通過穿戴式裝置偵測重訓時負重的方法。MuscleSense通過迴歸使用手臂或前臂上的穿戴式肌電流感測器的訊號來預測重訓室的負重。我們先收集了20個使用者的肌電流資料，然後去測試了不同位置的感測器對於我們的系統的影響。我們的結果顯示MuscleSense在預測間距為1公斤的負重時的使用線性迴歸支持向量機得到的方均根差為0.683公斤。

Strength training improves overall health, well-being, physical appearance, and sports performance.There are four major factors that affect training efficacy in a training session: exercise type, number of repetitions, movement velocity, and workload. Prior research has used wearable sensors to detect exercise type, number of repetitions, and movement velocity while training. However, detecting workload still requires instrumentation of exercise equipment such as exercise machines, or free weights. This paper presents MuscleSense, an approach that detects training weight through wearable devices.
In particular, MuscleSense uses various regressors to predicting weight using signals from wearable sEMG sensors mounted on user''s arm or forearm. We evaluated the effects of sensor placement and collected training data from 20 participants. The results from our user study show that MuscleSense achieves Root Mean Square Error(RMSE) of 0.683kg in sensing workload through sensors data from both forearm and arm using Support Vector Regressor of linear kernel.

誌謝i
摘要ii
Abstract iii
1 Introduction 1
2 Related Work 3
2.1 Training assisting system 3
2.2 Surface Electromyography and Force 4
3 Implementation 6
3.1 Device 6
3.2 Sensor Placement 7
3.3 Signal Processing and Smoothing 7
3.4 Machine Learning and Cross Validation 9
4 Field Study 11
5 User Study 12
5.1 Participants 12
5.2 Setup 12
5.3 Experimental Design 13
5.3.1Task 13
5.3.2Procedure 14
6 Results 16
7 Discussion 19
8 Conclusion 21
Bibliography 22

[1]F. Bai and C.-M. Chew. Muscle force estimation with surface emg during dynamicmuscle contractions: A wavelet and ann based approach. InEngineeringinMedicineand Biology Society (EMBC), 2013 35th Annual International Conference of theIEEE, pages 4589–4592. IEEE, 2013.
[2]I. Barofsky and M. W. Legro. Definition and measurement of fatigue.Reviews ofInfectious Diseases, 13(Supplement_1):S94–S97, 1991.
[3]V. Becker, P. Oldrati, L. Barrios, and G. Sörös. Touchsense: Classifying and measur-ing the force of finger touches with an electromyography armband. InProceedingsof the 9th Augmented Human International Conference, AH ’18, pages 34:1–34:3,New York, NY, USA, 2018. ACM.
[4]P. Bouissou, P. Estrade, F. Goubel, C. Guezennec, and B. Serrurier. Surface emgpower spectrum and intramuscular ph in human vastus lateralis muscle during dy-namic exercise.Journal of applied physiology, 67(3):1245–1249, 1989.
[5]L. Brody, M. T. Pollock, S. H. Roy, C. De Luca, and B. Celli. ph-induced effectson median frequency and conduction velocity of the myoelectric signal.Journal ofApplied Physiology, 71(5):1878–1885, 1991.
[6]M. Brzycki. Strength testing—predicting a one-rep max from reps-to-fatigue.Jour-nal of Physical Education, Recreation & Dance, 64(1):88–90, 1993.
[7]K.-H. Chang, M. Y. Chen, and J. Canny. Tracking free-weight exercises. InInter-national Conference on Ubiquitous Computing, pages 19–37. Springer, 2007.
[8]R. Chaudhri, J. Lester, and G. Borriello. An rfid based system for monitoring freeweight exercises. InProceedings of the 6th ACM Conference on Embedded NetworkSensor Systems, SenSys ’08, pages 431–432, New York, NY, USA, 2008. ACM.
[9]C. Choi, S. Kwon, W. Park, H.-d. Lee, and J. Kim. Real-time pinch force estimationby surface electromyography using an artificial neural network.Medicalengineering& physics, 32(5):429–436, 2010.
[10]J. M. Cissik. Basic principles of strength training and conditioning.NSCA’s Per-formance Training Journal, 1(4):7–11, 2002.
[11]C. J. De Luca. The use of surface electromyography in biomechanics.Journal ofapplied biomechanics, 13(2):135–163, 1997.
[12]C. J. De Luca. The use of surface electromyography in biomechanics.Journal ofapplied biomechanics, 13(2):135–163, 1997.
[13]H. Ding, L. Shangguan, Z. Yang, J. Han, Z. Zhou, P. Yang, W. Xi, and J. Zhao.Femo: A platform for free-weight exercise monitoring with rfids. InProceedingsof the 13th ACM Conference on Embedded Networked Sensor Systems, SenSys ’15,pages 141–154, New York, NY, USA, 2015. ACM.
[14]T. K. Evetovich. Progression models in resistance training for healthy adults (vol41, pg 687, 2009).Medicine and Science in Sports and Exercise, 41(6):1351–1351,2009.
[15]K. Häkkinen, P. V. Komi, M. Alén, and H. Kauhanen. Emg, muscle fibre and forceproduction characteristics during a 1 year training period in elite weight-lifters.Eu-ropean journal of applied physiology and occupational physiology, 56(4):419–427,1987.
[16]D. Hofmann, N. Jiang, I. Vujaklija, and D. Farina. Bayesian filtering of surface emgfor accurate simultaneous and proportional prosthetic control.IEEE Transactionson Neural Systems and Rehabilitation Engineering, 24(12):1333–1341, 2016.
[17]Y. Iravantchi, Y. Zhang, E. Bernitsas, M. Goel, and C. Harrison. Interferi: Gesturesensing using on-body acoustic interferometry. InProceedings of the 2019 CHIConference on Human Factors in Computing Systems, page 276. ACM, 2019.
[18]G. Kamen and G. E. Caldwell. Physiology and interpretation of the electromyogram.Journal of Clinical Neurophysiology, 13(5):366–384, 1996.
[19]R. Khurana, K. Ahuja, Z. Yu, J. Mankoff, C. Harrison, and M. Goel. Gymcam: De-tecting, recognizing and tracking simultaneous exercises in unconstrained scenes.Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Tech-nologies, 2(4):185, 2018.
[20]W. J. Kraemer and N. A. Ratamess. Fundamentals of resistance training: progressionand exercise prescription.Medicine and science in sports and exercise, 36(4):674–688, 2004.
[21]W. J. Kraemer, N. A. Ratamess, and D. N. French. Resistance training for health andperformance.Current sports medicine reports, 1(3):165–171, 2002.
[22]N. Li, Y. Hou, and Z. Huang. A real-time algorithm based on triaxial accelerometerfor the detection of human activity state. InProceedings of the 6th InternationalConference on Body Area Networks, BodyNets ’11, pages 103–106, ICST, Brussels,Belgium, Belgium, 2011. ICST (Institute for Computer Sciences, Social-Informaticsand Telecommunications Engineering).
[23]A. Luttmann, M. Jäger, and W. Laurig. Electromyographical indication of muscularfatigue in occupational field studies.InternationalJournalofIndustrialErgonomics,25(6):645–660, 2000.
[24]R. Merletti and P. Di Torino. Standards for reporting emg data.J ElectromyogrKinesiol, 9(1):3–4, 1999.
[25]F. Mobasser and K. Hashtrudi-Zaad. A comparative approach to hand force esti-mation using artificial neural networks.Biomedical engineering and computationalbiology, 4:BECB–S9335, 2012.
[26]A. Möller, L. Roalter, S. Diewald, J. Scherr, M. Kranz, N. Hammerla, P. Olivier,and T. Plötz. Gymskill: A personal trainer for physical exercises. InPervasiveComputing and Communications (PerCom), 2012 IEEE International Conferenceon, pages 213–220. IEEE, 2012.
[27]D.Morris, T.S.Saponas, A.Guillory, andI.Kelner. Recofit: Usingawearablesensorto find, recognize, and count repetitive exercises. InProceedings of the SIGCHIConference on Human Factors in Computing Systems, CHI ’14, pages 3225–3234,New York, NY, USA, 2014. ACM.
[28]K. Murao and T. Terada. A recognition method for combined activities with ac-celerometers. InProceedings of the 2014 ACM International Joint Conference onPervasive and Ubiquitous Computing: Adjunct Publication, UbiComp ’14 Adjunct,pages 787–796, New York, NY, USA, 2014. ACM.
[29]W. H. Organization. Global strategy on diet, physical activity and health.Retrieved April 10, 2019 from https://www.who.int/dietphysicalactivity/factsheet_adults/.
[30]M. Parai, P. D. Shenoy, S. Velayutham, C. K. Seng, C. Y. F. Yee, et al. Isometricmuscle strength as a predictor of one repetition maximum in healthy adult females:a crossover trial.Clinical Trials in Orthopedic Disorders, 1(2):71, 2016.
[31]S. Pizzolato, L. Tagliapietra, M. Cognolato, M. Reggiani, H. Müller, and M. Atzori.Comparison of six electromyography acquisition setups on hand movement classifi-cation tasks.PloS one, 12(10):e0186132, 2017.
[32]S. Reddy, M. Mun, J. Burke, D. Estrin, M. Hansen, and M. Srivastava. Using mobilephones to determine transportation modes.ACMTrans.Sen.Netw., 6(2):13:1–13:27,Mar. 2010.
[33]C. Seeger, A. Buchmann, and K. Van Laerhoven. myhealthassistant: A phone-basedbody sensor network that captures the wearer’s exercises throughout the day. InPro-ceedings of the 6th International Conference on Body Area Networks, BodyNets ’11, pages 1–7, ICST, Brussels, Belgium, Belgium, 2011. ICST (Institute for ComputerSciences, Social-Informatics and Telecommunications Engineering).
[34]M. H. Stone, D. Collins, S. Plisk, G. Haff, and M. E. Stone. Training principles:Evaluation of modes and methods of resistance training.Strength & ConditioningJournal, 22(3):65, 2000.
[35]P. Tomczykowska. The modern face of calisthenics. street workout as a new disci-pline of sport.Journal of Health Sciences, 3(11):011–020, 2013.
[36]J. Vredenbregt and G. Rau. Surface electromyography in relation to force, musclelength and endurance. InNew Concepts of the Motor Unit, Neuromuscular Disor-ders, Electromyographic Kinesiology, volume 1, pages 607–622. Karger Publishers,1973.