Dataset for auto identification of atmospheric disturbances and key parameter extraction based on all-sky airglow imaging observations of the Chinese Meridian Project

To meet the demand for efficient processing of massive airglow images from the Meridian Project, this study constructs an intelligent recognition and parameter extraction tool for atmospheric gravity waves and mesoscale traveling ionospheric disturbances based on machine learning techniques. A convolutional neural network classification model is used to screen images under clear night sky conditions, and the faster region-based convolutional neural network is applied to locate wave structures. For atmospheric gravity waves, a method based on two-dimensional Fourier transform is adopted to extract wavelength, propagation direction, and horizontal velocity, while for mesoscale traveling ionospheric disturbances, Canny edge detection combined with linear fitting is used to extract wave parameters.