臺灣博碩士論文加值系統

走路是日常生活中最平凡且最具功能性與重複性的運動，正常行走需要由神經系統、肌肉、骨骼及大小腦的高度協調，因此觀察步態可以為平衡障礙相關疾病，如巴金森氏病、中風等提供重要的線索。在現有的應用中，步態特徵通常由臨床醫師進行觀察，這些短期的行為觀察可能較為主觀，結果有時會不太一致，也不一定準確，特別是在臨床醫師缺乏經驗時。本論文目標是使用Kinect for Windows 感應器，開發一套便宜、易於使用，且能幫助病患及臨床醫師進行客觀評估分析及量化的系統，測驗內容採用一專門為臨床醫師所開發的簡易平衡評估系統測試（Mini Balance Evaluation Systems Test，Mini-BESTest）為動作標準，進行功能性的評估，透過設計之系統來分析量化動態步態測試的結果，並以Kinect深度鏡頭取得患者骨架相關資訊，通過頻率解調變及小波轉換進行時頻分析，計算步長、步頻及步速等步態參數，並藉由空間向量的投影及比對來判斷步態過程中的身軀穩定度與平衡，透過這些相關的參數來協助臨床醫師及病患量化整個步態測驗的過程及結果，增加其診斷的可信度與準確性，實際量測下，步數計算部分皆能準確算出，在步長計算時誤差皆在3cm以下。通過將步態測驗內容進行量化評估，不僅能節省醫患雙方的時間、提升效率，也能方便進行復健前後步態的比較或提供後續追蹤之參考。

Walking is the most common, functional and repetitive exercise in daily life. Normal walking requires a high degree of coordination between the nervous system, muscles, bones, cerebellum and brain. Therefore, observing the gait can be important clues of balance-related diseases, such as Parkinsons disease, stroke, etc. In existing applications, gait features are often observed by clinicians. These short-term behavioral observations may be subjective, and results may sometimes be inconsistent or inaccurate, especially when clinicians lack experience. The goal of this study is to use the Kinect for Windows sensor to develop a system that is cheap, easy to use, and can help patients and clinicians objectively evaluate, analyze, and quantify the examinations. The content of the test uses a simple balance assessment developed specifically for clinicians. Mini-BESTest is an action standard. It performs a functional evaluation. Through the designed system, it analyzes the results to quantify dynamic gait tests, and uses the Kinect depth lens to obtain a patients skeleton related information. Frequency demodulation and wavelet transforms are used for time-frequency analysis, and gait parameters such as step length, step frequency, and pace are calculated. The stability and balance of the body during the gait process are determined by the projection and comparison of space vectors. Related parameters can assist clinicians and patients in quantifying the process and results of the entire gait test, increasing the reliability and accuracy of its diagnosis. Under the actual measurement, the step number calculation part can accurately calculate the step length calculation that has an error is less than 3cm. By quantifying the content of the gait test, it will not only save the clinicians and patients time, and improve efficiency, but also facilitate the comparison of gait before and after rehabilitation or provide references for follow-up rehabilitation.

摘　要 i
ABSTRACT ii
誌　謝 iv
目　錄 v
圖目錄 vii
表目錄 ix
第一章 緒論 1
1.1 研究背景 1
1.2 研究目的 2
1.3 研究方法 3
1.4 論文架構 4
第二章 相關技術及運用探討 5
2.1 Kinect for Windows介紹 5
2.1.1 Kinect Version 2 6
2.1.2 深度測量技術 8
2.2 簡易平衡評估系統測試 10
2.3 灰色預測方法 14
第三章 系統架構與設計 18
3.1 系統架構 18
3.1.1 韌體架構 19
3.1.2 硬體架構 20
3.1.3 軟體架構 20
3.2 系統設計流程 22
3.3 Kinect步態評估系統設計 23
3.3.1 開發環境介紹 23
3.3.1.1 Visual Studio開發環境 23
3.3.1.2 系統介面 25
3.3.1.3 資料儲存 25
3.3.2 骨架預測補償 27
3.3.3 輸入訊號濾波 28
3.3.4 步頻計算 30
3.3.5 關節夾角 32
3.3.6 身軀擺動 33
3.3.7 臉部轉動角度 34
3.3.8 姿勢判斷 35
3.3.9 誤差量測 36
第四章 實驗結果 39
4.1 實驗環境與流程 39
4.2 實驗結果 41
4.2.1 改變速度 42
4.2.2 頭部水平旋轉 44
4.2.3 轉身停止 46
4.2.4 跨過盒子 48
4.2.5 同步任務計時 50
4.2.6 實驗結果評分 52
4.2.7 相關步態參數 52
第五章 結論與未來展望 54
5.1 結論 54
5.2 未來展望 55
參考文獻 56

[1]衛生福利部(焦點新聞)，https://www.mohw.gov.tw/cp-3160-25258-1.html
[2]衛生福利部統計處(死因統計)，https://dep.mohw.gov.tw/DOS/np-1776-113.html
[3]BESTest - Balance Evaluation Systems Test，http://www.bestest.us/
[4]Kinect - Wikipedia，https://en.wikipedia.org/wiki/Kinect
[5]Kinect - Windows 應用程式開發，
https://developer.microsoft.com/zh-tw/windows/kinect
[6]D. Scharstein and R. Szeliski, “High-accuracy stereo depth maps using structured light,” in Proc. of IEEE Computer Society Conf. on Computer Vision and Pattern Recognition, Washington, DC, USA, vol.1, pp.195-202, Jun. 2003.
[7]H. Sarbolandi, D. Lefloch and A. Kolb, “Kinect range sensing: structured-light versus time-of-flight Kinect,” Computer Vision and Image Understanding, vol. 139, pp.1-20, Oct. 2015.
[8]J. Zhu, L. Wang, R. Yang and J. Davis, “Fusion of time-of-flight depth and stereo for high accuracy depth maps,” in Proc. of IEEE Conference on Computer Vision and Pattern Recognition, Anchorage, Alaska, USA, pp.1-8, Jun. 2008.
[9]L. Yang, L. Zhang, H. Dong, A. Alelaiwi and A. E. Saddik, “Evaluating and improving the depth accuracy of kinect for windows v2,” IEEE Sensors Journal, vol. 15, no. 8, pp.4275-4285, Aug. 2015.
[10]Time-of-flight camera - Wikipedia，
https://en.wikipedia.org/wiki/Time-of-flight_camera
[11]C. Hsu and C. Chen, “Applications of improved grey prediction model for power demand forecasting” Energy Conversion and Management, vol. 44, no. 14, pp.2241-2249, Aug. 2003.
[12]M. Tabaszewski and C. Cempel, “Using a set of GM(1, 1) models to predict values of diagnostic symptoms,” Mechanical Systems and Signal Processing, vol. 52-53, pp.416-425, Feb. 2015.
[13]Microsoft Visual Studio - Wikipedia，
https://en.wikipedia.org/wiki/Microsoft_Visual_Studio
[14]I. W. Selesnick and C. S. Burrus, “Generalized digital Butterworth filter design,” IEEE Trans. on Signal Processing, vol. 46, no. 6, pp.1688-1694, Jun. 1998.
[15]L. G. Wiedemann, R. Planinc, I. Nemec and M. Kampel, “Performance evaluation of joint angles obtained by the kinect V2,” in Proc. of IET Int. Conf. on Technologies for Active and Assisted Living, London, UK, pp.1-6, Nov. 2015.
[16]B.C. Lee, A. Fung and T.A. Thrasher, “The effects of coding schemes on vibrotactile biofeedback for dynamic balance training in Parkinson’s Disease and healthy elderly individuals,” IEEE Trans. Neural Systems and Rehabilitation Engineering, vol. 26, no. 1, pp.153-160, Jan. 2018.
[17]H. A. Alabbasi, F. Moldoveanu and A. Moldoveanu, “Real time facial emotion recognition using Kinect v2 sensor,” IOSR Journal of Computer Engineering, vol. 17, no. 3, pp.61-68, May 2015.
[18]運用動態感測技術建立勞工上肢MSD危害評估研究，
https://labor-elearning.mol.gov.tw/base/10001/door/報告區/715_IOSH102-H321運用動態感應技術建立勞工上肢MSD危害評估研究.pdf
[19]S. Gaglio, G. L. Re and M. Morana, “Human activity recognition process using 3-d posture data,” IEEE Trans. on Human-Machine Systems, vol. 45, no. 5, pp.586-597, Oct. 2015.
[20]F. Gholami, D. A. Trojan, J. Kövecses, W. M. Haddad and B. Gholami, “A microsoft kinect-based point-of-care gait assessment framework for multiple sclerosis patients,” IEEE Journal of Biomedical and Health Informatics, vol. 21, no. 5, pp.1376-1385, Sep. 2017.
[21]X. Cai, G. Han, X. Song and J. Wang, “Single-camera-based method for step length symmetry measurement in unconstrained elderly home monitoring,” IEEE Trans. on Biomedical Engineering, vol. 64, no. 11, pp.2618-2627, Nov. 2017.
[22]M. A. Eltoukhy, C. Kuenze, J. Oh and J. F. Signorile, “Validation of static and dynamic balance assessment using microsoft kinect for young and elderly populations,” IEEE Journal of Biomedical and Health Informatics, vol. 22, no. 1, pp.147-153, Jan. 2018.