Synthetic generation of 2D data records based on Autoencoders

• DOI: 10.48550/arxiv.2502.13183
• Publication Date: 2025-02-18
• Authors: Darius Couchard; Oscar Javier Olarte; Rob Haelterman
• Keywords: FOS: Computer and information sciences; Computer Science - Machine Learning; Image and Video Processing (eess.IV); FOS: Electrical engineering, electronic engineering, information engineering; Electrical Engineering and Systems Science - Image and Video Processing; Machine Learning (cs.LG)

Gas Chromatography coupled with Ion Mobility Spectrometry (GC-IMS) is a dual-separation analytical technique widely used for identifying components in gaseous samples by separating and analysing the arrival times of their constituent species. Data generated GC-IMS typically represented as two-dimensional spectra, providing rich information but posing challenges data-driven analysis due to limited labelled datasets. This study introduces novel method generating synthetic 2D spectra using deep learning framework based on Autoencoders. Although applied here data, approach broadly applicable any spectral measurements where data are scarce. While performing component classification over dataset records, addition synthesized records significantly has improved performance, demonstrating method's potential overcoming limitations machine frameworks.