臺灣博碩士論文加值系統

由於現代科技的發達與健康意識的抬頭，不管是因為擔心各種體重過輕、過重所帶來的身體疾病或是健美、健身所必須的體重控制。體重儼然已經成為現今人們非常在意的一個身體訊號。
本研究的目的是希望藉由三軸加速規收集腿部訊號來觀察當體重變化時，步態特徵會有哪些規律的變化，並藉由這些變化來實現一個基於步態訊號的體重變化辨識系統。。
我們的受測者共有20位，男女各半，年齡介於21~50歲之間。我們請我們的受測者於雙腳腳踝上配戴我們的穿戴式三軸加速度感測模組，並赤腳來回行走於20公尺的磁磚地面長廊。受測者被規定須照著節拍器的頻率移動，重複實驗數次模擬真實體重變化。
我們所使用之步態週期精確值最早由Margareta Nordin等人於1989年所提出，他們將一個步態分割為七個事件，此定義於近代研究中已經被引用超過三千次。我們藉此精準的將每一步的開始結束標示出來，並切割出完整的訊號，並設計多種分類器比較分析哪種類型分類器更適合用來分析體重與步態的相關性。藉由特徵的比對，我們最後得到SVM分類器對於體重變化與步態特徵的相關性具有較高的辨識率與較快的運算效能。
總的來說，我們實踐了一個基於步態訊號的體重變化辨識系統，且此系統是準確且可靠的。未來更進一步的研究我們希望將技術應用在健身產業與老人遠端照護讓人類生活能更加智慧、舒適。

Due to the development of modern science and Health concept of the rise, Whether it is because of fear of overweight brought about by physical illness or fitness must be the weight control. Weight has become a body signal that people very concerned.
The purpose of this study is to use the triaxial acceleration to collect ankle signals to observe what changes in gait characteristics when weight changes .Through these changes to achieve a weight change identification system based on gait signal.
We have twenty subjects, ten men and ten women, aged between 21 to 50 years old. We asked our subjects wear wearable triaxial acceleration sensor on the ankle and walk back in a 20-meter tiles corridor. Subjects are required to walk according to the frequency of the metronome, repeat the experiment several times to simulate the real weight changes.
The exact definition of the gait cycle was first proposed by Margareta Nordin et in 1989. They divide a gait into seven events, which have been cited more than three thousand times in modern research. We use this mark the start and end of each step, and cut out the complete signal, then design a variety of classifiers to compare which classifier is more suitable for analysis of the correlation of weight change and gait features. By the comparison of the features, we finally get SVM classifier for the correlation of weight change and gait features has a high recognition rate and faster computing performance.
In general, we implemented a weight change identification system based on gait signals, and this system is accurate and reliable. In the future, we hope that the technology used in the fitness industry and the elderly medical care. So that life can be more intelligent and comfortable

摘要
ABSTRACT
誌謝
目錄
表目錄
圖目錄
第一章 緒論
第一節 研究背景
第二節 相關文獻
第三節 研究動機與目的
第二章 研究材料
第一節 步態週期與步態事件
第二節 訊號採集設備
第三節 步態實驗流程
第四節 符號定義
第三章 研究方法
第一節 訊號處理流程
第二節 濾波器設計
第三節 訊號分割
第四節 特徵擷取
第五節 分類器設計
第四章 實驗結果
第一節 驗證方法
第二節 分類器比較
第三節 SVM分析
第四節 Swing/Stance Phase分析
第五節 實驗結果小結
第五章 結論
參考文獻

1.Nordin, Margareta and Frankel, Victor Hirsch, Basic biomechanics of the musculoskeletal system, 1989.
2.D. J. Magee, Orthopedic Physical Assessment, Elsevier Health Sciences.
3.X. Li and S. J. Maybank and S. Yan and D. Tao and D. Xu, "Gait Components and Their Application to Gender Recognition," IEEE Transactions on Systems, Man, and Cybernetics, Part C (Applications and Reviews, pp. 145-155, 2008.
4.Stolze, H and Kuhtz-Buschbeck, J.P and Mondwurf, C and Boczek-Funcke, A and Jöhnk, K and Deusch, G and Illert, M, "Gait analysis during treadmill and overground locomotion in children and adults," Electroencephalography and Clinical Neurophysiology/Electromyography and Motor Control, pp. 490-497, 1997.
5.J, Osberg JS and McGinnis GE and DeJong G and Seward ML and Germaine, "Long-term utilization and charges among post-rehabilitation stroke patients.,," Am J Phys Med Rehabil, pp. 66-72, 1988.
6.Whittle, M. and Heinemann, B., Gait Analysis: An Introduction, 1991.
7.H. Zhang and J. Qian and L. Shen and Y. Zhang, "Research on healthy subject gait cycle phase at different walking speeds," 2012 IEEE International Conference on Robotics and Biomimetics, pp. 1349-1354, 2012.
8.Zhang, Yancheng Liu and Kun Lu and Songhua Yan and Ming Sun and D. Kevin Lester and Kuan, "Gait phase varies over velocities," Gait & Posture, pp. 756-760, 2014.
9.D., Robert J. Orr and Gregory, The Smart Floor: A Mechanism for Natural User, ACM Press, 2000.
10.Moe-Nilssen, Rolf and Helbostad, Jorunn L., "Estimation of gait cycle characteristics by trunk accelerometry," Journal of Biomechanics, pp. 121-126, 2004.
11.Grimson, L. Lee and W. E. L., "Gait analysis for recognition and classification," Proceedings of Fifth IEEE International Conference on Automatic Face Gesture Recognition, pp. 148-155, 2002.
12.Bacchi, Elisabetta and Bonin, Cecilia and Zanolin, Maria Elisabetta and Zambotti, Francesca and Livornese, Dario and Donà, Silvia and Tosi, Flavia and Baldisser, Giulia and Ihnatava, Tatsiana and Di Sarra, Daniela and Bonora, Enzo and Moghetti, Paolo, "Physical Activity Patterns in Normal-Weight and Overweight/Obese Pregnant Women," PLOS ONE, pp. 1-11, 2016.
13.Wang, J. S. and Lin, C. W.and Yang, Y. T. C and Ho, Y. J., "Walking Pattern Classification and Walking Distance Estimation Algorithms Using Gait Phase Information," IEEE Transactions on Biomedical Engineerin, pp. 2884-2892, 2012.
14.維基百科, "樸素貝葉斯分類器 --- 維基百科 自由的百科全書," 2017. [Online]. Available: https://zh.wikipedia.org/w/index.php?title=%E6%9C%B4%E7%B4%A0%E8%B4%9D%E5%8F%B6%E6%96%AF%E5%88%86%E7%B1%BB%E5%99%A8&oldid=44802549.
15.維基百科, "人工神經網絡 --- 維基百科 自由的百科全書," 2017. [Online]. Available: https://zh.wikipedia.org/w/index.php?title=%E4%BA%BA%E5%B7%A5%E7%A5%9E%E7%BB%8F%E7%BD%91%E7%BB%9C&oldid=43795095.
16.維基百科, "支持向量機 --- 維基百科 自由的百科全書," 2017. [Online]. Available: https://zh.wikipedia.org/w/index.php?title=%E6%94%AF%E6%8C%81%E5%90%91%E9%87%8F%E6%9C%BA&oldid=44135068.