DigitalGlobe to Map Buildings Using Machine Learning in the Cloud

DigitalGlobe’s WorldView 3 satellite captured this image of Sydney, Australia in January 2015. Photo: DigitalGlobe.

DigitalGlobe has formed a partnership with Ecopia Tech to use proprietary Artificial Intelligence (AI) algorithms and cloud computing to create building footprints. By using Ecopia’s U.S. Building Footprints powered by DigitalGlobe, customers will have current information on structures in their areas of interest.

Ecopia, a developer in DigitalGlobe’s Geospatial Big Data platform (GBDX) ecosystem, established a process to create building footprints quickly and at scale by leveraging machine learning in combination with DigitalGlobe’s cloud-based 100 petabyte imagery library. According to Ecopia, the service provides actionable insights for observing, analyzing, and monitoring business processes such as supply chain management, urban planning, and asset monitoring for industries that include energy, insurance, real estate, telecom, and location-based services. Starting with the United States, the two companies will extract 2D building footprints across the Earth, then refresh the datasets to find and track change over time.

DigitalGlobe and Ecopia intend to extract every building footprint in the United States by mid-year 2018, and many major international locations by the beginning of 2019. The service will be available off the shelf or on-demand, with or without the source high-resolution imagery, and will include the option to subscribe to regular updates.