[1O01-05] Nuclear Non-proliferation and Nuclear Security

10:45 AM - 11:00 AM JST(1:45 AM - 2:00 AM UTC)

[1O02] The development of detection and reasoning methods for nuclear material theft behaviors based on artificial intelligence and background analysis

*Zhan Li¹, Xingyu Song¹, Shi Chen¹, Kazuyuki Demachi¹ (1. The University of Tokyo)

Keywords:

Nuclear security,Human malicious behaviors identification,Artificial intelligence,Background analysis

A novel detection and reasoning method for nuclear material theft behaviors is proposed in this research. Based on our previous research, computer vision technologies are used to extract the human action sequences for further reasoning process. For the more complex scenarios of nuclear material theft, an extra tree-structured analysis process is proposed to determine the status of background objects. Besides, this background analysis process is well integrated in our previously proposed explainable reasoning methods. The results show that our proposed methods prove to be promising to detect the behaviors of nuclear material theft.