Deakin University
Browse

File(s) under permanent embargo

Skeleton-free task-specific rapid upper limb ergonomie assessment using depth imaging sensors

conference contribution
posted on 2017-01-05, 00:00 authored by Darius Nahavandi, Mohammed Hossny
© 2016 IEEE. Rapid upper body assessment (RULA) is, undoubtedly, one of the most frequently used ergonomic scoring sheets in different industries. The scoring is based on measuring joint angles during assembly tasks. However, the process is labour intensive and requires ergonomists to study hundreds of hours of videos to identify musculoskeletal disorders associated with different assembly tasks. In this paper we propose automating the process using depth imaging sensors such as Kinect and a random decision forest (RDF). The main contribution in this paper is relaxing the need to construct a skeleton and train the RDF on the whole posture at once at the price of training different RDFs for different tasks. The accuracy of the proposed method converges to 95%.

History

Event

IEEE Sensors. Conference (2016 : Florida, Orlando)

Publisher

IEEE

Location

Florida, Orlando

Place of publication

Piscataway, N.J.

Start date

2016-10-30

End date

2016-11-02

ISSN

1930-0395

eISSN

2168-9229

ISBN-13

9781479982875

Language

eng

Publication classification

E1 Full written paper - refereed

Copyright notice

2016, IEEE

Title of proceedings

2016 Conference Proceedings of IEEE Sensors