Research on a Visual Electronic Nose System Based on Spatial Heterodyne Spectrometer

Area of Science:

• Spectroscopy
• Gas Sensing Technology
• Electronic Nose Systems

Background:

• Light absorption gas sensing offers parallelism and responsiveness for electronic noses (e-nose).
• Spatial heterodyne spectrometer (SHS) provides hyperspectral resolution, converting gas absorption spectra into 2D interferograms for image processing.

Purpose of the Study:

• To propose and verify a visual e-nose system based on SHS.
• To explore the feasibility of using image processing techniques for gas analysis with SHS data.

Main Methods:

• Constructed a theoretical model for the visual e-nose system.
• Acquired experimental visual maps of gas absorption spectra.
• Employed Local Binary Pattern (LBP) and Gray-Level Co-occurrence Matrix (GLCM) for feature extraction.
• Utilized distance similarity algorithms (Correlation Coefficient, Euclidean Distance to Centroids) for pattern recognition.

Main Results:

• Generated visual maps from experimental data.
• Successfully extracted features using LBP and GLCM.
• Demonstrated pattern recognition capabilities using classification algorithms.

Conclusions:

• The proposed visual e-nose system based on SHS is feasible for gas sensing.
• Image processing techniques combined with SHS show promise for e-nose applications.
• The study validates the use of LBP, GLCM, and distance similarity metrics for gas pattern recognition.