[1E01-03] Nuclear Security

[1E02] Detection of Insiders Malicious Behaviors for Nuclear Security by Hand Motion Analysis

*Shi Chen¹, Kazuyuki Demachi¹, Yusuke Kawasaki¹, Yutaro Nakashima¹ (1. Univ. of Tokyo)

Keywords:Malicious Behaviors Detection, Hand Motion Tracking, Kinect, Stacked Auto-Encoder

After Fukushima Daiichi nuclear power plant accident, the importance of nuclear security is increased. Especially as the threat to nuclear power plants, sabotage is worthy of attention. In addition, when considering the sabotage, due attention should be paid to insiders. Moreover, hand motion is an important part of human activity and it has high contribution to high-accuracy detection of insider malicious behaviors.
In this research, the real-time hand motion detection system was developed by using Kinect v2. Also, the possibility of malicious behavior detection was explored by using Stacked Auto-Encoder.