A simulation system for decision of camera position and angle: Human motion analysis in case of dangerous posture scenarios

In the manufacturing industry, while automation by machines advances to save labor, there are many situations where workers with high technical skills are needed to create added value for products. However, it is necessary to predict accidents during hands-on operations and reduce human errors by human motion analysis. In this paper, we propose a simulation system for detecting dangerous postures based on human motion analysis. We compare the simulation results with human motion analysis considering video images captured by the camera. The experimental results show that the proposed system is an effective system for deciding the position and angle of the camera. • In general, a network system uses multiple cameras to capture the whole human body movements. In this paper, we utilize a monocular camera to get human motion data. • The proposed system can decide the camera position and angle for human motion analysis. • Possible use cases for the proposed system include visualizing the imaging results of human body parts and presenting this information to the operator of the camera-based system to assist in determining the camera position.