Previous | Table of contents | Next
Developments in computational and data science of machine learning and AI are changing conventional methods including remote sensing and data analysis using Earth observation data.
Many of the conventional analyses are based on models, requiring understanding of numerous complex polynomials, knowledge of sensors and remote sensing and analysis techniques.
In addition, higher sensor resolution and longer observation periods made it difficult to process large amounts of data , and such analysis have been limited to disaster response and target study areas.
However, even without specialized knowledge, machine learning and AI made analysis possible as data mining.
For example, as machine learning and AI can process the complex analysis of parameters and models as image analysis, even researchers without knowledge of remote sensing learn to use Earth observation data.
In addition to the increase of the researchers, the analysis scope of another observation periods and areas will be expanded as well. It is also expected to use data that has not been used before.
Furthermore, new applications of Earth observation data including prediction and super-resolution are being studied, and expectations for the use of Earth observation data are growing.
Thus, the use of machine learning and AI is a significant opportunity for both researchers and Earth observation data providers, and is expected to develop in the future.
CEOS agencies have also conducted studies of the analysis using machine learning and AI, and have presented these studies including in WGISS.
These contents and findings are summarized in this document to assist researchers and data providers as they begin their research.