A Spectral-Angle-Loss Variational Autoencoder Deep Embedding (Sa-VaDE) Model for Clustering FORS Spectra of Micro-metallurgical remains

OSF Preprints (OSF Preprints) 2026 Preprint

Authors: Siran Liu
Cited by: 0

Abstract

In this study, feature compression and cluster analysis were performed on the 3963 FORS spectra of micro-remains. The model used is the Sa-VaDE model which is improved on the basis of VaDE model, the validity of the Sa-VaDE model is confirmed by analyzing the final clustering results.

Fields of Study & Subjects

Mineral Processing and Grinding Image Processing and 3D Reconstruction Iron and Steelmaking Processes

Publication details

Journal / source
OSF Preprints (OSF Preprints)
Publication date
10 March 2026
Document type
Preprint