A fast image recognition method of mobile point radioactive source based on sequential bayesian analysis

Abstract

In the fields of astronomical observation, homeland security, medical diagnosis and arms control verification, the monitoring and imaging of mobile point radioactive sources is common. In many cases, the number of its rays entering the detector is small and may be even smaller than that of the background signals. Traditional radiation image recognition requires accumulating a certain number of detection events before recognition, which is not suitable to the fast recognition of the above mobile point radioactive sources. In this paper, a fast image recognition method of the mobile point radioactive source based on the sequential Bayesian analysis is proposed. The method can analyze the positions of the ray particles in the radiation image one by one according to the time sequence, and fast determine the existence of the mobile point radioactive source. First, the theoretical framework of the method is built. Second, the feasibility of the method is analyzed by means of numerical simulation. The simulation results show that, the method is suitable to the image recognition of the mobile point radioactive sources with different motion states. At last, the practical application scenarios of the method are analyzed. The calculation results show that, the method is suitable to the image recognition of the mobile point radioactive sources under the conditions of extremely low signal-to-noise ratios and waiting for the ray signals of the radioactive sources for a long time.