Image processing of a spectrogram produced by Spectrometer Airglow Temperature Imager

The Spectral Airglow Temperature Imager is an instrument, specially designed for investigation of the wave processes in the Mesosphere-Lower Thermosphere. In order to determine the kinematics parameters of a wave, the values of a physical quantity in different space points and their changes in the time should be known. An approach for image processing of registered spectrograms is proposed. A detailed description is made of the steps of this approach, related to recovering CCD pixel values, influenced by cosmic particles, dark image correction and filter parameters determination.

💡 Research Summary

The paper presents a comprehensive image‑processing workflow tailored for the Spectral Airglow Temperature Imager (SATI), an instrument specifically built to investigate wave processes in the mesosphere‑lower thermosphere (MLT). SATI records two‑dimensional spectrograms of atmospheric airglow on a CCD detector, where each pixel corresponds to a narrow wavelength band emitted by O₂ at typical altitudes around 80‑90 km. Accurate extraction of temperature and wave parameters from these spectrograms requires meticulous correction of several sources of systematic error and noise.

The authors first address the contamination of raw CCD data by high‑energy cosmic particles (cosmic rays) and radioactive decay events, which manifest as isolated spikes or “hot pixels.” They implement a statistical outlier detection scheme based on a 5‑sigma deviation from the local neighborhood median. Identified outliers are replaced by a weighted average of the eight surrounding pixels, and pixels that repeatedly exhibit spikes across successive frames are flagged as permanently defective and excluded from further analysis.

Next, the paper details dark‑frame correction. CCDs generate a temperature‑ and exposure‑time‑dependent dark current that adds a spatially varying background to every image. To remove this bias, the authors acquire a series of dark frames under identical exposure and temperature conditions, compute a master dark by averaging at least thirty frames, and linearly interpolate between dark frames taken before and after each science exposure to account for slow temporal drift. Subtracting this interpolated dark level from the raw spectrogram yields a background‑free image ready for quantitative analysis.

The third step concerns the determination of filter parameters. SATI’s optical system includes a fixed interference filter and a diffraction grating that together define the instrument’s spectral response. Manufacturing tolerances and temperature fluctuations cause the central wavelength (λ₀) and bandwidth (σ) of the filter to shift slightly. The authors obtain laboratory calibration spectra, fit each spectral line with a Gaussian model I(λ)=I₀·exp