Classification of gaseous UF6 assay by femtosecond LIBS in the 424.4 nm spectral region using numerical HOGSVD-DTW features

Abstract

This technical note presents experimental results using numerical features of fs-LIBS data to classify the assay value of a gaseous UF₆ material. The data-driven feature vectors are computed by Higher Order Generalized Singular Value Decomposition (HOGSVD) and Dynamic Time Warp (DTW). The method achieves 96.97% accuracy in spectral classification testing with fs-LIBS samples obtained from a UF₆ material with five known assay values ranging from 0.287% to 61.740%, with 100% accuracy for the four largest assay values ranging from 4.615% to 61.740%.