A novel approach to the localization and the estimate of radioactivity in contaminated waste packages via imaging techniques

Dismantling nuclear power plants entails the production of a large amount of contaminated (or potentially contaminated) material whose disposal is of crucial importance. Most of the end products have to be stored in special repositories but it can happen that some of them are slightly contaminated or not contaminated at all, making it possible to free release them. One possible approach to free release measurements uses Large Clearance Monitors, chambers surrounded by plastic scintillation detectors that allow a decision about the clearance of waste packages up to 1000 kg. Due to the composite nature of the detectors in a Large Clearance Monitor, it is easy to imagine that one can apply 3D imaging algorithms to localize radioactive sources inside a waste package. In this work we will show how a special algorithm that maximizes the conditional informational entropy allows decisions about the clearance of portions of the sample.