Title :
Study of artificial neural network on explosive detection with PFTNA method
Author :
Wang, Haidong ; Li, Yuanjing ; Yang, Yigang ; Hu, Shuzhi ; Chen, Boxian ; Gao, Wenhuan
Author_Institution :
Dept. of Eng. Phys., Tsinghua Univ., Beijing, China
Abstract :
This work describes a study of artificial neural network on explosive detection. With PFTNA method, gamma-ray spectra of some explosives and common materials were acquired. A multilayer perception neural network was designed and trained with these spectra. Back-propagation algorithm was used as the training method. For faster training, different improved algorithms were tested. Good results have been obtained with some algorithms.
Keywords :
backpropagation; gamma-ray detection; multilayer perceptrons; PFTNA method; artificial neural network; back-propagation algorithm; explosives detection; gamma-ray spectra; multilayer perception neural network; Artificial neural networks; Composite materials; Explosives; Gamma ray detection; Gamma ray detectors; Multi-layer neural network; Neural networks; Neutrons; Pulse generation; Scattering;
Conference_Titel :
Nuclear Science Symposium Conference Record, 2005 IEEE
Print_ISBN :
0-7803-9221-3
DOI :
10.1109/NSSMIC.2005.1596295
