Abstract
Fast nuclide identification with higher accuracy rate is a significant requirement within nuclear applications. So effective spectrum identification algorithms are developed for low resolution spectra of NaI(TI) scintillator and Si(Li) detectors. Identification of X-ray and gamma complex spectra is the main objective of current research. So the experimental gamma and X-ray data are trained and tested using artificial neural network (ANN), support vector machine (SVM) and similarity classifiers. The features of acquired X-ray and gamma spectra are extracted using eight algorithms. These algorithms depend on fusion of time-domain descriptors (FTDD), electromyography (EMG), multiscale wavelet packet (MWP), multiscale wavelet packet with statistics (MWPS), principle component analysis (PCA), multi-dimensional scale, the preserved linear projection (LPP) and diffusion map. Robustness of these algorithms is investigated in terms of noise degradations such as Gaussian, Rician, Rayleigh and other complex degradations. Classification accuracy is investigated with the source name. The recognized spectrum is analyzed from view of peak width calibration, efficiency calibration, sum peak analysis, peak-to-Compton ratio (PCR). The rapid identification process is conducted with the algorithm based on electromyography method for both gamma and X-ray spectra. However, the algorithm based on diffusion map realizes the slowest spectrum identification. It is concluded that, for gamma and X-ray spectra, the SVM classifier achieves the fastest identification with maximum rate of 99%. Finally, the ANN is observed to achieve better rate of 100% with slower identification process depending on FTDD, EMG, LPP methods. The proposed approach helps the realization of fastest spectrum identification and classification of gamma and X-ray spectra within nuclear applications with higher robustness and accuracy.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.