Machine learning for localization of radioactive sources via a distributed sensor network

Abstract In this paper, we focus on the detection and localization of radioactive sources by exploiting supervised machine learning. Machine learning is utilized in a wide variety applications due to its effectiveness prediction autonomous decision-making. However, applying would only be effective when representative features for application can acquired, through which algorithms trained. Hence, first, present feature extraction technique source localization, then propose parameter estimation method via A distributed sensor network employed assist estimating source’s location intensity. We that evaluates vector using reading position each located region where radiation detected. The based data fusion process, single value provided represent both coordinates corresponding given sensor. After extraction, apply decision tree regression localize source. To examine proposed work, performance comparison carried out with recent existing methods terms accuracy execution time. Experimental results show algorithm provides accurate intensity achieves good compromise between