[3P12-13] Technology for Nuclear Non-proliferation and Nuclear Security

[3P13] Development of Vision-based Insider Sabotage Detection System for Nuclear Security

*Shi Chen¹, Kazuyuki Demachi¹ (1. The University of Tokyo)

Keywords:Nuclear security, Physical protection system, Insider sabotage, Deep neural network

Fukushima Daiichi nuclear power plant accident raised a fear that terrorists could cause a similar accident by acts of sabotage against nuclear facilities and the importance of nuclear security increased after this accident. Especially as a threat to nuclear facilities, sabotage by insider is worthy of attention. It appears that hand motion has high contribution to human activity and a significant portion of sabotage behaviors can be detected through hand motion monitoring.
In this study, a vision-based Deep Neural Network (DNN) model is proposed for hand motion recognition. Based on this model, a real-time hand motion detection system will be developed and the possibility of insider sabotage detection will be explored in response to the certain limitations of physical protection system (PPS) in nuclear facilities.