An AI Approach in Radioactive Source Localization by a Network of Small Form Factor CZT Sensors

Abstract

We present a small form factor (0.5𝑐𝑚3) static CZT sensor net-work consisted of a number of Non- Directional Detectors (NDD)capable to localize a stationary radiation source in 3D. The localiza-tion is performed with a fusion algorithm based on AI techniques.The algorithms are based on Multilayer Perseptron Neural Net-work (MLP) and Gradient Boosted Decision Trees (BDTG). Theyhave been trained using simulated data produced by the SWORDsimulation software based on Geant4 framework. The localizationefficiency of the algorithms was verified with experimental datataken in our laboratory using a137𝐶𝑠source of180𝜇𝐶𝑖. The local-ization resolution of the order of 10cm to 15cm has been archivedin Vertical and Horizontal directions respectively and of the orderof less than 20cm in the depth direction within a monitored volumeof 5m x 2.8m x 2m