Advertisement

Automated Video Lifting Posture Classification Using Bounding Box Dimensions

  • Runyu Greene
  • Yu Hen Hu
  • Nicholas Difranco
  • Xuan Wang
  • Ming-Lun Lu
  • Stephen Bao
  • Jia-Hua Lin
  • Robert G. Radwin
Conference paper
Part of the Advances in Intelligent Systems and Computing book series (AISC, volume 820)

Abstract

A method is introduced for automatically classifying lifting postures using simple features obtained through drawing a rectangular bounding box tightly around the body on the sagittal plane in video recordings. Mannequin postures were generated using the University of Michigan 3DSSPP software encompassing a variety of hand locations and were classified into squatting, stooping, and standing. For each mannequin posture a rectangular bounding box was drawn tightly around the mannequin for views in the sagittal plane and rotated by 30 º horizontally. The bounding box dimensions were measured and normalized based on the standing height of the corresponding mannequin. A classification and regression tree algorithm was trained using the height and width of the bounding box to classify the postures. The resulting algorithm misclassified 0.36% of the training-set cases. The algorithm was tested on 30 lifting postures collected from video recordings a variety of industrial lifting tasks, misclassifying 3.33% of test-set cases. The sensitivity and specificity, respectively were 100.0% and 100.0% for squatting, 90.0% and 100.0% for stooping, and 100.0% and 95.0% for standing. The algorithm was capable of classifying lifting postures based only on dimensions of bounding boxes which are simple features that can be measured automatically and continuously. We have developed computer vision software that continuously tracks the subject’s body and automatically applies the described bounding box.

Keywords

Computer vision Musculoskeletal disorders Exposure assessment 

References

  1. 1.
    Chen CH, Hu YH, Yen TY, Radwin RG (2013) Automated video exposure assessment of repetitive hand activity level for a load transfer task. Hum Fact 55(2):298–308CrossRefGoogle Scholar
  2. 2.
    Akkas O, Lee CH, Hu YH, Yen TY, Radwin RG (2016) Measuring elemental time and duty cycle using automated video processing. Ergonomics 59(11):1514–1525CrossRefGoogle Scholar
  3. 3.
    Greene RL, Azari DP, Hu YH, Radwin RG (2017) Visualizing stressful aspects of repetitive motion tasks and opportunities for ergonomic improvements using computer vision. Appl ErgonGoogle Scholar
  4. 4.
    ACGIH (American Conference of Governmental Industrial Hygienists). TLV®/BEI® Introduction (2017) http://www.acgih.org/tlv-bei-guidelines/tlv-bei-introduction
  5. 5.
  6. 6.
    Breiman L, Friedman J, Stone CJ, Olshen RA (1984) Classification and regression trees. CRC PressGoogle Scholar
  7. 7.
    Bao S, Howard N, Spielholz P, Silverstein B (2006) Quantifying repetitive hand activity for epidemiological research on musculoskeletal disorders–Part II: comparison of different methods of measuring force level and repetitiveness. Ergonomics 49(4):381–392CrossRefGoogle Scholar
  8. 8.
    Lu M, Waters T, Krieg E, Werren D (2014) Efficacy of the revised NIOSH lifting equation for predicting risk of low back pain associated with manual lifting: a one-year prospective study. Hum Fact 56(1):73–85CrossRefGoogle Scholar
  9. 9.
    Safetyvideopreviews (2012) Manual Material Handling/Safe Lifting. https://www.youtube.com/watch?v=rrI2n8qehrY&t=8s
  10. 10.
    University of Michigan Center for Ergonomics. 3DSSPP: Background Information (2017) https://c4e.engin.umich.edu/tools-services/3dsspp-software/3dsspp-background-information/
  11. 11.
    University of Michigan Center for Ergonomics (2014) Paper Flopping - Job Modification. https://www.youtube.com/watch?v=61cu5qvH0kM&index=54&list=PLn5IJRj74S88rnFFV6ObxS6nFdDXUFiGW
  12. 12.
    University of Michigan Center for Ergonomics (2017) Stacking, Facing Line2 CE. https://www.youtube.com/watch?v=MxTgvuhVAJA&t=55s

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  • Runyu Greene
    • 1
  • Yu Hen Hu
    • 1
  • Nicholas Difranco
    • 1
  • Xuan Wang
    • 1
  • Ming-Lun Lu
    • 2
  • Stephen Bao
    • 3
  • Jia-Hua Lin
    • 3
  • Robert G. Radwin
    • 1
  1. 1.University of Wisconsin-MadisonMadisonUSA
  2. 2.National Institute for Occupational Safety and HealthCincinnatiUSA
  3. 3.Washington Department of Labor and IndustriesOlympiaUSA

Personalised recommendations