Brass Material Analysis with Deep-Learning-Based CdTe Semiconductor X-ray Fluorescence System

Ajin Jo, Wonho Lee

Research output: Contribution to journalArticlepeer-review

Abstract

An X-ray fluorescence (XRF) imaging system is a material analysis system that can represent the material distribution and types of elements by detecting characteristic X-rays emitted from each element. A CdTe semiconductor detector array whose detection efficiency is significantly higher than a silicon drift detector (SDD) is utilized with a deep learning method to improve the energy spectral analysis. In this study, deep learning models for material discrimination and quantitation were applied based on 20,000 energy spectra obtained from Fe, Ni, Cu, and Zn rod phantoms, and a brass phantom was analyzed to verify that Cu, Zn, and brass can be distinguished from each other and that the amount of Cu and Zn in each phantom can be quantitatively analyzed.

Original languageEnglish
JournalIEEE Transactions on Nuclear Science
DOIs
Publication statusAccepted/In press - 2022

Keywords

  • CdTe semiconductor detector array
  • Convolutional neural network
  • Data models
  • Deep Learning
  • Deep learning
  • Imaging
  • Phantoms
  • Training
  • X-ray Fluorescence (XRF) imaging
  • X-ray imaging
  • Zinc

ASJC Scopus subject areas

  • Nuclear and High Energy Physics
  • Nuclear Energy and Engineering
  • Electrical and Electronic Engineering

Fingerprint

Dive into the research topics of 'Brass Material Analysis with Deep-Learning-Based CdTe Semiconductor X-ray Fluorescence System'. Together they form a unique fingerprint.

Cite this