Neutron-gamma discrimination method based on blind source separation and machine learning

作     者：Hanan Arahmane El-Mehdi Hamzaoui Yann Ben Maissa Rajaa Cherkaoui El Moursli Hanan Arahmane;El-Mehdi Hamzaoui;Yann Ben Maissa;Rajaa Cherkaoui El Moursli

作者机构：Universite´Paris-SaclayCEAList91120 PalaiseauFrance Equipe des Sciences de la Matie`re et du Rayonnement(ESMaR)Faculty of SciencesMohammed V University in RabatB.P.1014 RPRabatMorocco National Centre for Nuclear EnergyScience and Technology(CNESTEN)B.P.1382 R.P.10001RabatMorocco Laboratory of TelecommunicationsNetworks and Service SystemsNational Institute of Posts and TelecommunicationsAllal El Fassi AvenueRabatMorocco 

出 版 物：《Nuclear Science and Techniques》 (核技术（英文）)

年 卷 期：2021年第32卷第2期

页      面：70-80页

核心收录：

学科分类：12[管理学] 082704[工学-辐射防护及环境保护] 1201[管理学-管理科学与工程(可授管理学、工学学位)] 081104[工学-模式识别与智能系统] 08[工学] 0827[工学-核科学与技术] 0835[工学-软件工程] 0811[工学-控制科学与工程] 0812[工学-计算机科学与技术（可授工学、理学学位）] 

基　　金：L’Ore´al-UNESCO for the Women in Science Maghreb Program Grant Agreement No.4500410340 

主　　题：Blind source separation Nonnegative tensor factorization(NTF) Support vector machines(SVM) Continuous wavelets transform(CWT) Otsu thresholding method 

摘      要：The discrimination of neutrons from gamma rays in a mixed radiation field is crucial in neutron detection *** approaches have been proposed to enhance the performance and accuracy of neutron-gamma ***,their performances are often associated with certain factors,such as experimental requirements and resulting mixed *** main purpose of this study is to achieve fast and accurate neutron-gamma discrimination without a priori information on the signal to be analyzed,as well as the experimental ***,a novel method is proposed based on two *** first method exploits the power of nonnegative tensor factorization(NTF)as a blind source separation method to extract the original components from the mixture signals recorded at the output of the stilbene scintillator *** second one is based on the principles of support vector machine(SVM)to identify and discriminate these *** addition to these two main methods,we adopted the Mexican-hat function as a continuous wavelet transform to characterize the components extracted using the NTF *** resulting scalograms are processed as colored images,which are segmented into two distinct classes using the Otsu thresholding method to extract the features of interest of the neutrons and gamma-ray components from the background *** subsequently used principal component analysis to select the most significant of these features wich are used in the training and testing datasets for ***-variance analysis is used to optimize the SVM model by finding the optimal level of model complexity with the highest possible generalization *** this framework,the obtained results have verified a suitable bias–variance trade-off *** achieved an operational SVM prediction model for neutron-gamma classification with a high true-positive *** accuracy and performance of the SVM based on the NTF was evaluated and validated by comparing it to the charge comparison