A tale of two big geospatial big data breakthroughs: quantifying built environments

Approximately three years ago, we came in contact with PSMA Australia. Though a relatively small organization, PSMA had an audacious goal: To map the entirety of Australia in tremendous detail. The PSMA team believed a highly accurate and detailed map of the entire continent’s built environment would bring immense value to numerous customers across a variety of industries.

PSMA soon realized the only way to build a map of this magnitude was by partnering with a satellite imagery company with sufficient coverage and granularity to observe and record buildings. Additionally, PSMA required the extraction of building footprints as a GIS-ready dataset, and that extraction had to be performed quickly, accurately and cost-effectively.

Fast forward. Today we are on the cusp of completing this ambitious objective with our partners at PSMA. PSMA’s product Geoscape is currently in the market and has generated significant interest and excitement both within Australia and across the globe. PSMA harnessed the power of our high-resolution satellite imagery, GBDX (our cloud-based analytics platform), and incredibly efficient artificial-intelligence algorithms to build the first, single authoritative dataset at continental scale. This breakthrough has piqued the interest of organizations around the world looking to develop an accurate and comprehensive understanding of the built environment.

The excitement about our partnership with PSMA to jointly develop and launch Geoscape has caught the attention of customers across multiple industries including insurance, civil government, energy, telecommunications and global development organizations. All these customers are trying to answer similar questions:

  • How many buildings do we have?
  • Where are they?
  • What are they?

 

We’ve been contacted by energy companies interested in detecting and extracting building footprints around their pipelines to manage and mitigate their risk profile with confidence. County governments have seen the press and reached out to us to learn how machine learning methods applied to satellite imagery can be used to build a uniform, consistent database of all the structures within their administrative boundaries. Insurance organizations have expressed interest in building footprints to determine properties and structures at risk of flood and fire. The list of applications goes on!

DigitalGlobe building footprints, powered by GBDX, in partnership with Ecopia
Key to our story is Ecopia Tech, a company in Waterloo, Canada we met when we started our conversations with PSMA. At the time, Ecopia was two guys and a dog in a garage. But they had developed very attractive technology that focused their AI-powered algorithms on detecting and extracting beautiful polygons out of satellite imagery. (Beauty is in the eye of the beholder. Being a GIS nerd at heart, these are beautiful footprints.)

We quickly saw the potential of these footprints for multiple markets. We struck up a partnership with Ecopia to begin testing the market and assessing where we could provide value to customers. Over time we have integrated our technologies and business models, and now we are launching a joint product offering.

Building on the groundbreaking vision and work completed down under, DigitalGlobe with our partners at Ecopia are proud to announce the global launch of our building footprint product—derived from our best-in-class satellite imagery and using our analytics platform, GBDX.

The future is bright where Artificial Intelligence intersects with high-resolution satellite imagery and a scalable cloud-based analytics platform. At DigitalGlobe, we have many future-focused projects underway, collaborating with partners and customers to drive innovation with emerging technological advancements.

Interested in working with us to help solve difficult problems? Dive into the full PSMA story and learn about how our partnership with Ecopia led to the launch of our AI-enabled building footprints product.