275 KB, PDF document
Research output: Contribution to conference - Without ISBN/ISSN › Abstract › peer-review
Research output: Contribution to conference - Without ISBN/ISSN › Abstract › peer-review
}
TY - CONF
T1 - Super Resolution Hyperspectral Imaging based Automated Inspection of Nuclear Fuel Pellets
AU - Zabalza, Jaime
AU - Parker, Andrew
AU - Bandala Sanchez, Manuel
AU - Murray, Paul
AU - Marshall, Stephen
AU - Ma, Xiandong
AU - Taylor, C. James
AU - Joyce, Malcolm
PY - 2022/11/7
Y1 - 2022/11/7
N2 - This work presents a new methodology for the automated inspection of nuclear fuel pellets based on Single Image Super Resolution (SISR) and Hyperspectral Imaging (HSI). HSI technology provides optical images in which the pixels contain comprehensive spectral information, normally hundreds of channels (wavelengths) covering the Visible Near InfraRed (VNIR) region in the electromagnetic spectrum. Therefore, the spectral information provided by HSI can be used for inspecting images of pellets pixel-wise. However, the spatial resolution in HSI is lower in comparison to conventional imaging, and SISR is proposed for enhancing the HSI images. Results showed how techniques such as Principal Component Analysis (PCA) can be applied to SR-HSI images to effectively exploit the HSI spatialspectral content and generate maps for the automated detection of potential abnormalities on the surface of nuclear fuel pellets. While experiments used color chalk as analogues of PWR pellets, results with sintered UO2 will be presented at the conference.
AB - This work presents a new methodology for the automated inspection of nuclear fuel pellets based on Single Image Super Resolution (SISR) and Hyperspectral Imaging (HSI). HSI technology provides optical images in which the pixels contain comprehensive spectral information, normally hundreds of channels (wavelengths) covering the Visible Near InfraRed (VNIR) region in the electromagnetic spectrum. Therefore, the spectral information provided by HSI can be used for inspecting images of pellets pixel-wise. However, the spatial resolution in HSI is lower in comparison to conventional imaging, and SISR is proposed for enhancing the HSI images. Results showed how techniques such as Principal Component Analysis (PCA) can be applied to SR-HSI images to effectively exploit the HSI spatialspectral content and generate maps for the automated detection of potential abnormalities on the surface of nuclear fuel pellets. While experiments used color chalk as analogues of PWR pellets, results with sintered UO2 will be presented at the conference.
M3 - Abstract
T2 - 2022 IEEE Nuclear Science Symposium
Y2 - 5 November 2022 through 12 November 2022
ER -