A Semi-Automatic System for Posture Risk Assessment

摘要

RULA (Rapid Upper Limb Assessment) is a survey method that assesses biomechanical and postural loading on the whole body with particular attention to the neck, trunk and upper limbs. In this paper we present a semi-automatic posture recognition system, to collect data and to evaluate human posture risk based on RULA. The proposed system uses two synchronized video cameras to capture images or posture samples of the operator performing its job. RULA scores are then automatically computed from these posture samples, based on semi-automatic and automatic image processing algorithms. These algorithms can extract visual features from the human upper limb, and obtain its model. The resulting scoring generates an action list which indicated the level of intervention required to reduce the risks of injury due to physical loading on the operator. The developed system is intended to be used as a tool for ergonomic investigations of workplaces.