Paper
27 October 2023 Infrared spectroscopy electronic instruments combined with lightweight DAResNet12 network for gas identification method research
Zhikang Wang, Guodong Zhao, Junming Zhang
Author Affiliations +
Proceedings Volume 12922, Third International Conference on Electronics, Electrical and Information Engineering (ICEEIE 2023); 1292216 (2023) https://doi.org/10.1117/12.3008609
Event: The Third International Conference on Electronics, Electrical and Information Engineering (ICEEIE 2023), 2023, Xiamen, China
Abstract
In order to quickly and accurately identify the gas infrared spectral components, a relatively new research result, the extended residual network (DA-ResNet12), is introduced. The gas infrared spectral data measured using an infrared spectroelectronic meter were selected for the experiment. First, the raw data are preprocessed, and the spectral features are extracted, and the feature vectors are generated by feature engineering; then, the lightweight DA-ResNet12 model is built and the model parameters are adjusted; finally, the proposed lightweight DA-ResNet12 model is compared with the commonly used algorithms according to the evaluation indexes such as the classification accuracy, the required training time, and the size of the model parameters. The experimental results demonstrate that the newly proposed DA-ResNet12 network model outperforms other comparative algorithms in terms of the number of model parameters, accuracy and time efficiency.
(2023) Published by SPIE. Downloading of the abstract is permitted for personal use only.
Zhikang Wang, Guodong Zhao, and Junming Zhang "Infrared spectroscopy electronic instruments combined with lightweight DAResNet12 network for gas identification method research", Proc. SPIE 12922, Third International Conference on Electronics, Electrical and Information Engineering (ICEEIE 2023), 1292216 (27 October 2023); https://doi.org/10.1117/12.3008609
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Convolution

Infrared radiation

Feature extraction

Infrared spectroscopy

Data modeling

Thermal modeling

Education and training

Back to Top