DeepEmSat: Deep Emulation for Satellite Data Mining

Abstract

The growing volume of Earth science data available from climate simulations and satellite remote sensing offers unprecedented opportunity for scientific insight, while also presenting computational challenges. One potential area of impact is atmospheric correction, where physics-based numerical models retrieve surface reflectance information from top of atmosphere observations, and are too computationally intensive to be run in real time. Machine learning methods have demonstrated potential as fast statistical models for expensive simulations and for extracting credible insights from complex datasets. Here, we develop DeepEmSat: a deep learning emulator approach for atmospheric correction, and offer comparison against physics-based models to support the hypothesis that deep learning can make a contribution to the efficient processing of satellite images.

Deep learning for processing and analysis of remote sensing big data: a technical review

“…Xu, Cervone, Franch, and Salvador (2020) presented an artificial intelligence/deep learningbased solution to characterize the atmosphere at different vantage points and to retrieve target spectral properties. A deep learning emulator approach for atmospheric correction was developed, suggesting that a deep learning model can be trained to emulate a complex physical process (Duffy et al, 2019).…”

Meta-reservoir computing for learning a time series predictive model of wind power

“…However, it does not give hints on building a more general model. Deep models encounter overfitting issues (Duffy et al, 2023). As the environment changes, the prediction by deep models deviates from the real value and needs to be modified by using data from the new environment.…”