In this episode, I talk with Steve Brumby, co-founder, CEO and CTO of Impact Observatory, about sustainability and environmental risk analysis. Impact Observatory uses satellite imagery and machine learning to empower decision-makers with planetary insights.

Highlights:

  • Using machine learning to generate thematic maps to represent land cover and land use.
  • Geospatial data from the European Space Agency’s Copernicus program that is available on a variety of platforms.
  • The importance of identifying the relevant output for end users and others in the value chain.
  • How machine learning engineers sometimes discover things used by remote sensing scientists that are no longer necessary.
  • Keeping models simple.
  • Mitigating bias in models by using large and globally diverse datasets.
  • Get to know your customer and their pain points, then craft a machine learning solution that works for them – if you’re lucky, it’ll also work for others.
  • Finding the things you’re passionate about – both the technology and helping the customers in that space.

Quotes:

“Impact Observatory, as the name suggests, is designed to give decision makers all around the world access to the sort of data that actually the US government and the big firms on Wall Street now currently have access to.”

“Impact Observatory is designed to use satellite imagery and artificial intelligence, which for this audience we can be a bit more precise, to use deep learning algorithms that look at the spatial, spectral, and temporal information from times sequences of observations from orbit, and use that to understand how the world is changing in near time. And deliver that to decision makers in very easy to consume ways so that you can address direct questions that decision makers have. Like how much deforestation is happening in the watershed for my city, or there’s a whole lot of people growing crops in some region and are they following sustainable agricultural practiceslike even just simple ones like crop rotation?”

“You can be a machine learning scientist and you can come up with a fantastic algorithm, and your work is not done because you have to produce outputs in a format that really ties into the workflow of the customer.”

“You have to get to know the customer, find out where their pain points are, and be connecting those dots because there’s way more problems in the world than there are solutions. So there’s plenty of opportunity, for people who are trained up on the latest types of machine learning, there is a gazillion opportunities for you to find a real pain point that would have measurable impact if somebody solved it.”

Links:


Resources for Computer Vision Teams:

LinkedIn – Connect with Heather.
Computer Vision Insights Newsletter – A biweekly newsletter to help bring the latest machine learning and computer vision research to applications in people and planetary health.
Computer Vision Strategy Session – Not sure how to advance your computer vision project? Get unstuck with a clear set of next steps. Schedule a 1 hour strategy session now to advance your project.
Foundation Model Assessment – Foundation models are popping up everywhere – do you need one for your proprietary image dataset? Get a clear perspective on whether you can benefit from a domain-specific foundation model.