Convolutional neural networks achieve over 80% accuracy in meteor and meteorite recognition, suggesting advancements in machine learning for astronomy.
In this paper, we present a convolutional neural network (CNN)-based architecture trained on a dataset of meteorites and terrestrial rocks and another dataset trained on meteors and light sources. For meteorites, the dataset comprises augmented images from the meteorite collection at the Sharjah Academy for Astronomy, Space Sciences, and Technology (SAASST). For meteors, the images are taken from the United Arab Emirates (UAE) Meteor Monitoring Network (MMN). Such a project’s significance is to expand machine learning applications in astronomy to include the solar system’s small bodies upon contact with the Earth’s atmosphere. This, in return, acts as deep learning research, which examines a computer’s ability to mimic a human’s brain in recognizing meteorites from rocks, and meteors from airplanes and other noise sources. When testing the CNN models, results have shown that both the meteorite and meteor models reached an accuracy of above 80%.
No takes yet. Share an insight, caveat, or question.
Alowais et al. (2023) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: