Deep learning enables automated extraction of building footprints and road networks from satellite imagery

Automated feature extraction from satellite imagery has made major progress in the last year. Accurate building footprints extracted from high resolution satellite imagery are becoming available from companies such as Ecopia, which has just announced a partnership with DigitalGlobe, whose satellites are capable of 30 cm (approximately one foot) resolution. Also NVIDIA has demonstrated the ability to automate detection of many road networks using sophisticated algorithms and multi-spectral high resolution imagery.

Last year a competition organized by DigitalGlobe, CosmiQ Works, and NVIDIA challenged developers to create algorithms that extract building footprints from DigitalGlobe's high resolution (30 cm) multi-band imagery made publicly available on SpaceNet. The winner of the competition applied a deep neural network model developed originally for automated medical image analysis. The algorithm was very successful in identifying building footprints in Las Vegas including suburban residential and odd shaped commercial and public buildings. It was also successful in identifying large standalone buildings in Shanghai, Paris and Khartoum, but areas where there were many adjoining buildings with little or no separation remain challenging. The solutions offered by the developers in the competition were sufficiently successful that the organizers concluded that the winning algorithms achieved performance with potential for automated mapping tasks such as keeping maps up-to-date and assisting first responders during natural disasters. The source code for the winning implementations was made available on the SpaceNetChallenge GitHub repository.

DigitalGlobe has just announced a partnership with Ecopia, which has established an automated process to create building footprints quickly and at scale by leveraging DigitalGlobe's Geospatial Big Data platform (GBDX) and advanced machine learning in combination with DigitalGlobe’s imagery library. The two companies plan to automatically extract accurate 2D building footprints globally, then refresh the datasets periodically to find and track changes over time. These datasets would be valuable to municipal governments for permitting purposes and for first responders after disasters.  Ecopia is developing a database for all of Australia which will include every building with a roof area greater than 9 square meters. The database will contain building footprints for about 15 million buildings. The database includes linkages to other datasets including geocoded address, property data and administrative boundaries. Ecopia intend to update the dataset regularly to be able to track changes, especially urban sprawl.

NVIDIA has released the results of several deep learning algorithms that illustrate just how difficult identifying road networks is and the sophistication of the available tools. These include using multi-band spectral imagery to identify the material properties of the road surface itself (asphalt, gravel, packed earth) and using more sophisticated algorithms. Applied to four cities across the globe, the results are excellent for Las Vegas. Outside North America, however, automatically identifying road networks was found to be more challenging. NVIDIA utilized high performance GPU compute resources provided by the NVIDIA GPU Cloud and Amazon Web Services.

Read more