2020 Convention session – Generalized Planetary Remote Sensing
This session on generalized planetary remote sensing aired live at the 2020 virtual CGIAR Convention on Big Data in Agriculture.
Numerous global challenges require a globally comprehensive observation of many variables simultaneously. Combining satellite imagery with machine learning presents an opportunity for assembling such observations. Current solutions require custom systems, extensive expert knowledge, access to imagery, and major computational resources to estimate a single variable using regional or global imagery.
The speaker of this session presents a new, general solution to constructing global observations with machine learning, where a single method for transforming satellite imagery is sufficiently descriptive to predict nearly any ground-level variables that are recoverable through inspection of a satellite image. Following the presentation, discussants discuss the potential of this approach in agricultural applications.
- Esther Rolf: Electrical Engineering and Computer Sciences Department, UC Berkeley, Ph.D. Student
- Tamma Carleton: Bren School of Environmental Science, UC Santa Barbara, Assistant Professor Environmental Economics, Climate Change
- Jawoo Koo: Senior Research Fellow, IFPRI, Geospatial Data CoP Lead, CGIAR Platform for Big Data in Agriculture
November 19, 2020