This episode is sponsored by Bee Maps.
Dr. Robin Cole joins Alastair to discuss the application of deep learning to satellite imagery and the growing community in this field. Robin shares his background in optical physics and how he transitioned into remote sensing and deep learning. Deep learning, a specific part of the AI family, uses neural networks to process satellite imagery and has seen significant advancements since 2012. The volume of satellite imagery requires automated means of analysis and indexing, and deep learning models have made this possible. There are standardized methods and libraries, such as PyTorch and TorchGeo, that facilitate deep learning in the geospatial domain. The community Robin has built focuses on sharing resources, knowledge, and expertise in deep learning and Earth observation.
Relevant links:
On the Geomob podcast every week we discuss themes from the geo industry, interview Geomob speakers, and provide regular updates about our own projects.
The Geomob podcast is hosted by:
Host Ed Freyfogle
Ed Freyfogle, co-founder of OpenCage
Host Steven Feldman
Steven Feldman, of KnowWhere Consulting
Host Alastair Graham
Alastair Graham, freelance geospatial/earth observation consultant.