An Ensemble Approach to Computationally Efficient Radiological Anomaly Detection and Isotope Identification

Radiological source search is a challenging task involving detection and identification of weak sources in constantly changing radiological background. As now, many algorithms have been proposed; however, their computational complexity and, hence, reliance on power intensive processing units inhibit low-power applications systems. In this work, we introduce the anomaly filter (AF) algorithm; computationally light, yet effective time-series algorithm based exponential weighted moving average (EWMA) Poisson deviance statistics. Then, demonstrate that proposed can be used ensemble with other more to achieve both increased performance reduced consumption. The AF were thoroughly benchmarked against several existing algorithms. results show outperforms conventional algorithms, approach improves overall isotope Furthermore, non-negative matrix factorization approach-based (NMF-ID) combined implemented single-board microcontroller, consumption was measured. This NMF-ID almost by factor 100, while improving system.