應用Kinect體感器 於全身肌肉骨骼危害評估

Assessment of Whole-Body Musculoskeletal Disorders by using Kinect Sensor

孫培峰(Pei-Feng Sun)

指導教授 ： 田方治

國立臺北科技大學/管理學院/工業工程與管理系碩士班/碩士(2014年)

https://doi.org/10.6841/NTUT.2014.00203

摘要

鑒於近年來影像感應技術的進步，以拍攝方式即可建立3D模組或是進行動態追蹤，其中由Microsoft所推出的Kinect體感器，可藉由拍攝之體動作進行遊戲操作，在操作過程中不需配帶任何感應裝置即可追蹤骨架並判斷姿勢之功能，將其擴展至人因領域，應用於肌肉骨骼危害評估之中。
傳統肌肉骨骼危害檢核表中，對於所分析之作業必須以人眼長時間觀察判斷，需耗費大量的人力時間。以Kinect取代人力之優勢，除了可自動化進行作業評估外，在拍攝完畢後即可產出報表，並且評估頻率取決於鏡頭拍攝速度，對於Kinect每秒約30fps之速度，可評估動作快速以人眼較難判斷之作業，且以Kinect拍攝可達到標準化危害等級之效果，利於建立各職類肌肉骨骼危害資料庫。
將此類攝影裝置應用於自動化姿勢判斷可建立自動化肌肉骨骼危害評估系統，而在目前Kinect硬體上仍有許多限制，因此本研究主要以The European Assembly Worksheet(EAWS)作為危害評估工具所開發之自動評估系統，並分析其運用之效果。實驗方面分為靜態以及動態實驗，從兩種不同實驗研究Kinect之硬體限制對於自動工作危害評估的影響。

關鍵字

Kinect ； 姿勢分析 ； 可攜式動作捕捉設備

並列摘要

With the rapid development of depth image sensor technology, we can create 3D model or dynamic tracking posture by using kinect, which is develop by Microsoft for Xbox360. It make for game controlling by using human body.It can track the skeleton and determine the position of the joint without wear any sensors. To extend it to human factors, applied to musculoskeletal hazard assessment.
Traditional musculoskeletal hazards checklist , for the job must be analyzed in order to determine the long observation of the human eyes, need to spend a lot of manpower time. Kinect to replace manpower and automation of evaluating. While shooting is complete report can be output, and assess the frequency depends on the speed of shots about 30fps. Kinect can achieve the effect of shooting with a standardized level of hazard. Conducive to the establishment of the various categories of musculoskeletal hazards database.
The body sensing devices used in automation judge posture can create automated hazard assessment of musculoskeletal system. In the current Kinect hardware still has many limitations, this study use EAWS tool for musculoskeletal hazard assessment. Testing automation assessment with Kinect hardware limitations.

並列關鍵字

Kinect ； Posture analysis ； Portable motion capture equipment

參考文獻

21. Richard Zhi-Ling Hu, Adam Hartfiel, James Tung, Adel Fakih, Jesse Hoey and Pascal Poupart, “3D Pose Tracking of Walker Users’ Lower Limb with a Structured-Light,” Computer Vision and Pattern Recognition Workshops (CVPRW), 2011 IEEE Computer Society Conference on, Colorado Springs, CO, 20-25 June 2011, Pages 29 - 36.

29. P. Benavidez and M. Jamshidi, “Mobile Robot Navigation and Target Tracking System,” System of Systems Engineering (SoSE), 2011 6th International Conference on, Albuquerque, NM, 27-30 June 2011, Pages 299 - 304.

30. Andrea Sanna, Fabrizio Lamberti, Gianluca Paravati and Federico Manuri, “A Kinect-based natural interface for quadrotor control,” Entertainment Computing, Volume 4, Issue 3, August 2013, Pages 179–186.

1. Tilak Dutta, "Evaluation of the Kinect™ sensor for 3-D kinematic measurement in the workplace,” Applied Ergonomics, Volume 43, Issue 4, July 2012, Pages 645–649.

2. H. Gonzalez-Jorge, B. Riveiro, E. Vazquez-Fernandez, J. Martínez-Sánchez and P. Arias, “Metrological evaluation of Microsoft Kinect and Asus Xtion sensors,” Measurement, Volume 46, Issue 6, July 2013, Pages 1800–1806.