Video-Based Ergonomic Repetitive Hand Motion Analysis

Details

• Personal Author:
  Lee C-H
• Description:
  Hand activity analysis for industrial works that require intensive repetitive hand motions provides important information to facilitate ergonomic analysis of the hand activity levels (HAL). The outcome may lead to a sound work-rest schedule with significant health benefits. The hand activity analysis consists of two important parts: (a) tracking the hand motion trajectory, and (b) determining whether the hand is exerting force or resting. In this research, we develop novel techniques to determine the state of repetitive hand activities (exerting forces or resting) based on kinematic information such as hand motion trajectory. In a repetitive hand task, the percentage of a cycle that the hand is holding a tool, or a product is called a duty cycle. Force is exerted by the hand during the duty cycle. Accurate estimation of the duty cycle provides the required information for exposure assessment in evaluating repetitive hand work. A key hypothesis of this research is that the motion trajectory and associated kinematic measurements (speed, acceleration) can provide sufficient information for duty cycle estimation. To validate this hypothesis, we developed a machine-learning algorithm using hand motion kinematic measurements extracted from videos of both simulated laboratory and practical factory repetitive hand work environments. We further propose a computer vision method to automatically measure wrist flexion and extension from a 2D video for occupational health and safety research. This algorithm tracked skeletal joints of the elbow, wrist, and hand to estimate the wrist flexion/extension angle between the hand and forearm. Based on the estimated angles, wrist posture was classified as flexion (palmar bending), neutral (no bending), or extension (dorsal bending) for each cycle of hand movement. Applying to a set of laboratory videos of a simulated repetitive hand motion task and selected video frames of hand flexion instances from industrial field video data, we demonstrated the feasibility of using this algorithm for assessing the state of hand activities during manual work. In summary, we demonstrated two non-intrusive, automated observation methods allowing long-term, large-scale collection of exposure data. They facilitate repetitive hand motion analysis for monitoring occupational health and safety and can help prevent job-related hand injuries. [Description provided by NIOSH]
• Subjects:
  Automation Ergonomics Occupational Health Safety
• Keywords:
  Activity Levels Kinesiology Motion Studies Repetitive Work Simulation Methods
• Publisher:
  University of Wisconsin
• Document Type:
  Text
• Funding:
  Grant
• Genre:
  Dissertation
• Place as Subject:
  OSHA Region 5 ; Wisconsin
• CIO:
  National Institute for Occupational Safety and Health (NIOSH)
• Topic:
  Workplace Safety & Health
• Location:
  North America
• Pages in Document:
  1-61
• NIOSHTIC Number:
  nn:20068908
• Citation:
  Madison, WI: University of Wisconsin, 2021 Aug; :1-61
• Federal Fiscal Year:
  2021
• Performing Organization:
  University of Wisconsin, Milwaukee
• Peer Reviewed:
  False
• Start Date:
  20160901
• Source Full Name:
  Video-based ergonomic repetitive hand motion analysis
• End Date:
  20190831
• Collection(s):
  National Institute for Occupational Safety and Health
• Main Document Checksum:
  urn:sha-512:c7ffce70afdfaaf46748bd5dc43f7f8bf24aebe6b2985d44860a13d32678abfab76239a05ef8d4d420e39fd5118ba025c45a3cf586f75a23dc499ca0f646e7cd
• Download URL:
  https://stacks.cdc.gov/view/cdc/207471/cdc_207471_DS1.pdf
• File Type:
  [PDF - 1.60 MB ]