Development of a new approach for rapid identification and classification of uranium ore concentrate powders using textural and spectroscopy signatures

A new approach for rapid identification of UOC powders using colour, textural and spectroscopy signatures was developed using a sample dataset consisting of 79 different industrial UOC powders. In particular, in this study, a machine learning approach was used to develop supervised texture- and spectroscopy-based classification models for each colour category of uranium ore concentrates. UOC SEM images were collected at different magnifications. A total of 524 texture features describing sample morphology were extracted from each image using 7 textural feature extraction algorithms. Hyperspectral images of the samples were collected in the wavelength range 900-1700 nm and the mean spectrum of each sample was calculated, giving the spectral features of the UOCs. The texture and spectral features were used as input to the machine learning approach which consisted of a feature reduction step and a classification step to provide the texture- and spectroscopy-based models. An average prediction accuracy of 82% and 97% were obtained for the texture- and spectral-based classification models, respectively, on hold-out test sets. This new approach enables rapid classification of unknown uranium ore concentrate powders using only their textural and hyperspectral images. By combining the classification results from the two models, the most probable production process and chemical composition can be identified, and the provenience of the seized sample can be narrowed down to a few matching possibilities.

FONGARO Lorenzo;  FUTSAETHER Cecilia;  TOMIC Oliver;  LANDE Isak Biringvad;  KVAAL Knut;  WALLENIUS Maria;  MAYER Klaus; 

2023-10-11

ELSEVIER

JRC132119

0169-7439 (online),   

https://www.sciencedirect.com/science/article/pii/S0169743923001089?via%3Dihub,    https://publications.jrc.ec.europa.eu/repository/handle/JRC132119,   

10.1016/j.chemolab.2023.104858 (online),