Posture recognition of nuclear power plant operators by supervised learning

This paper proposes a postures recognition method of nuclear power plant operators by a supervised learning approach. Operator´s silhouettes in the images are detected by combinations of several image processing techniques such as a background subtraction, noise reductions and others. Their postures are recognized by a machine learning technique. Their operations are summarized and visualized with human body computer graphics. The posture recognition is a challenging task because an operator usually takes various postures during power plant operations. To recognize the detected silhouettes, the method uses the four postures that have been classified by the cognitive scientists engaged in human factors research of nuclear power plant operations. In evaluation experiments with over twenty thousand images, the silhouettes are classified to the four postures successfully and the operations are summarized by the classified postures.