Multi-label Classification with Optimal Thresholding for Multi-composition Spectroscopic Analysis
This work addresses gas species identification in cluttered environments for spectroscopic analysis, representing an incremental improvement.
The paper tackled the problem of identifying gas species in multi-gas mixtures using multi-label neural networks with optimal thresholding, achieving superior performance over conventional methods when signal-to-noise ratio and training sample size were adequate.
In this paper, we implement multi-label neural networks with optimal thresholding to identify gas species among a multi gas mixture in a cluttered environment. Using infrared absorption spectroscopy and tested on synthesized spectral datasets, our approach outperforms conventional binary relevance - partial least squares discriminant analysis when signal-to-noise ratio and training sample size are sufficient.