9081 images dataset for: Detection of weather events in optical satellite data using deep convolutional neural networks

The dataset used for classifying satellite images of different weather events, including tropical cyclones, convective cell clouds, convective roll clouds, wildfires, and dust storms, as presented in Li, Y., & Momen, M. (2021).

Anyone can use this dataset by citing the following paper:
Li & Momen (2021): Detection of weather events in optical satellite data using deep convolutional neural networks. Remote Sensing Letters, 12(12), 1227–1237. https://doi.org/10.1080/2150704X.2021.1978581.

This dataset consists of 5 image datasets, one for each class 2202 for Cell Convection, 1655 for Dust Storms, 2147 for Tropical Cyclones, 1037 for Roll Convections and 2040 for Wild Fires, totaling 9081 optical satellite images. These data were processed and classified using deep convolutional neural networks (InceptionV3, ResNet50, VGG16, and VGG19) for accurate and automatic weather pattern detection.