Development of a radiative transfer model for the determination of toxic gases by Fourier transform–infrared spectroscopy with a support vector machine algorithm

We propose a hard expectation-maximization-based normalized matched filter (EM-NMF) for the detection of chemical warfare agent (CWA) clouds under background contamination. The NMF, which is one of the most powerful detectors, requires background statistics calculated from a background training dataset. However, in practice, because the training dataset is likely to contain CWA-on background pixels, the performance of the NMF is severely degraded. This phenomenon is referred to as background contamination. To address this issue, we propose an algorithm that estimates the posterior probability of each pixel belonging to either the background or the CWA class. The optimal posterior probabilities are obtained by maximizing the loglikelihood of a contaminated dataset using the EM algorithm. Based on the posterior probability, we extract CWA-free background pixels from the contaminated dataset and design a hard EM-NMF with extracted CWA-free background pixels. We demonstrate that the proposed algorithm is an effective solution for background contamination, via experimental results conducted with actual CWA data measured by a Bruker HI-90 instrument in an outdoor setting as well as synthetic CWA data.

show abstract

“…The test statistic of the NMF is obtained by substituting (9) and (11) into the likelihood ratio of x [14], as follows:…”

Section: A Nmf For Remote Cwa Cloud Detectionmentioning

confidence: 99%

Design of a Hard Expectation-Maximization-Based Normalized Matched Filter (EM-NMF) for the Detection of Chemical Warfare Agents Under Background Contamination

Lee

Park

et al. 2020

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

Self Cite

View full text Add to dashboard Cite

show abstract

“…The SVM trained with this preprocessed data achieved a higher performance than the correlation coefficients after removing background signatures [ 15 ], adaptive subspace detectors [ 16 ], and the SVM trained with preprocessed data in previous studies [ 13 ]. Nam et al [ 17 ] classified gases using an SVM, discerning the presence or absence of classified gases based on SVM scores. Kim et al [ 18 ] utilized deep neural networks and convolutional neural networks, outperforming the SVM as a comparative model.…”

Section: Introductionmentioning

confidence: 99%

Semi-Supervised Autoencoder for Chemical Gas Classification with FTIR Spectrum

Jang,

Kwon,

Nam

et al. 2024

Sensors

Self Cite

View full text Add to dashboard Cite

Chemical warfare agents pose a serious threat due to their extreme toxicity, necessitating swift the identification of chemical gases and individual responses to the identified threats. Fourier transform infrared (FTIR) spectroscopy offers a method for remote material analysis, particularly in detecting colorless and odorless chemical agents. In this paper, we propose a deep neural network utilizing a semi-supervised autoencoder (SSAE) for the classification of chemical gases based on FTIR spectra. In contrast to traditional methods, the SSAE concurrently trains an autoencoder and a classifier attached to a latent vector of the autoencoder, enhancing feature extraction for classification. The SSAE was evaluated on laboratory-collected FTIR spectra, demonstrating a superior classification performance compared to existing methods. The efficacy of the SSAE lies in its ability to generate denser cluster distributions in latent vectors, thereby enhancing gas classification. This study established a consistent experimental environment for hyperparameter optimization, offering valuable insights into the influence of latent vectors on classification performance.

show abstract

“…However, the performance of such detection algorithms is signi cantly affected by the database of distinctive gas spectra and the background information for eliminating signals other than the gas spectrum. Recently, machine learning detection algorithms have been applied by leveraging the nonlinearity of the algorithm and statistics driven by a massive amount of data, such as support vector machines [23] and deep neural networks [24]. Such approaches have improved the detection capability compared to probabilistic models; however, obtaining high-quality and large quantities of the training dataset under various conditions is challenging.…”

Section: Introductionmentioning

confidence: 99%

Effect of the dual library on long-range (3 km) SF 6 (Hexafluoride Sulfur) detection efficiency by portable Stand-off FTIR (Fourier Transform Infrared) detector

Byun

Nam

Kim

2022

Preprint

Self Cite

View full text Add to dashboard Cite

A critical limitation in developing portable Fourier transform-infrared (FTIR) stand-off chemical detectors is the detection library optimization, which reduces the number of false alarms and maintains the detection performance. Ideally, a detection library should be established through high-resolution (0.5 cm-1) FTIR spectroscopy in the laboratory. However, owing to the weight requirement of a portable FTIR stand-off detector, its optical system and mirror-moving range have limitations. Consequently, most portable FTIR stand-off detectors have been developed with a low resolution of approximately 4–16 cm-1. In this study, we developed a portable FTIR stand-off detector with a dual library of SF6. Moreover, we investigated its detection efficiency effect depending on the single library condition through realistic long-range (3 km) SF6 spectrum data in a field test.

show abstract

Development of a radiative transfer model for the determination of toxic gases by Fourier transform–infrared spectroscopy with a support vector machine algorithm

Cited by 11 publications

References 8 publications

Design of a Hard Expectation-Maximization-Based Normalized Matched Filter (EM-NMF) for the Detection of Chemical Warfare Agents Under Background Contamination

Design of a Hard Expectation-Maximization-Based Normalized Matched Filter (EM-NMF) for the Detection of Chemical Warfare Agents Under Background Contamination

Semi-Supervised Autoencoder for Chemical Gas Classification with FTIR Spectrum

Effect of the dual library on long-range (3 km) SF 6 (Hexafluoride Sulfur) detection efficiency by portable Stand-off FTIR (Fourier Transform Infrared) detector

Contact Info

Product

Resources

About