Machine Learning-Based Atmospheric Phenomena Detection Platform
Abstract
Satellite remote sensing is essential in advancing the state of knowledge of the Earth's surface and atmosphere processes. Data and imagery from these platforms is produced at rates that challenge the scalability and efficiency of classification methods currently employed by the science community. Machine learning techniques can address these computational issues however the uniqueness of each platform and phenomena often results in specialized post processing analysis software for each application. The Phenomena Portal is an attempt to consolidate multiple Machine Learning models applied to Earth and atmosphere remote sensing datasets into a single visualization and analysis platform. The tool serves as an event database resource for scientists who seek to leverage the spatiotemporal knowledge provided for additional applications. We will demonstrate the development of models applied to multiple remote sensing platforms and their integration into the Portal. In addition to feature specific visualization, generalized spatiotemporal analysis for all phenomena is highlighted. Expansion of the available phenomena and near-real time detection and analysis functionality development efforts are ongoing.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2019
- Bibcode:
- 2019AGUFMIN53C0754G
- Keywords:
-
- 1906 Computational models;
- algorithms;
- INFORMATICS;
- 1916 Data and information discovery;
- INFORMATICS;
- 1942 Machine learning;
- INFORMATICS;
- 1956 Numerical algorithms;
- INFORMATICS