Portal de Periódicos da CAPES

Real-Time Monitoring of the Posture at the Workplace Using Low Cost Sensors

2018; Springer Nature; Linguagem: Inglês

10.1007/978-3-319-96083-8_85

2194-5357

Marco Tarabini, Marco Marinoni, Matteo Mascetti, Pietro Marzaroli, Francesco Corti, Hermes Giberti, Paolo Mascagni, Alberto Villa, Tammy Eger,

Musculoskeletal pain and rehabilitation

The aim of this paper is to show a method that can be used to monitor the human posture in an industrial environment. The method is based on the fu- sion of the data coming from two different sensors: a time-of-flight camera (Microsoft Kinect V2) and a wearable motion capture system that uses inertial measurement units to identify the body posture (Notch Wearable). The combined use of these two systems overcomes the intrinsic limitations of the two methods, deriving from occlusions and electromagnetic interferences, respectively. First, the algorithms implemented and the calibration of the two measurement systems in a controlled environment are described. Second, the method applied in a workplace to monitor the posture of the workers during different tailoring operations, is explained. The data acquired have been analyzed in the time domain, and used to compute the cumulative probability density function of different body angles. The results are compared to the subjective evaluation of occupational doctors, and used to compute the OCRA index in an auto- mated way, for the assessment of workers exposure to repetitive movements of the upper limbs.