Использование численного моделирования переноса излучения для решения задачи дистанционного обнаружения источника гамма-квантов с заданным спектром
Year: 2012
Type: article
Abstract: Within the framework of mathematical methods applied to pattern recognition, a procedure for detection of a gamma-ray source with a given spectrum was developed and verified. The procedure involves usage of learning neural networks for separation into two classes (background and effect plus background) of spectra, obtained by means of gamma-spectrometer. Verification was carried out by means of a numerical simulation of radiation transport with Monte-Carlo method. Real conditions of an experiment were simulated, a set of spectra was obtained that was used for learning of multilayer perseptron. Verification was carried out separately for background in order to obtain a probability for false alarm and separately for the source with background in order to estimate a probability for detection of that source.
Language: Russian
Subfield: Radiation
Field: Physics and Astronomy
Domain: Physical Sciences
Open Access status: diamond
OpenAlex ID: W376579832
