Menu Close

Microsoft’s new SpaceEye technology enables imaging satellites to ‘see’ through clouds using AI

Last year, Microsoft announced Azure Space, making the powerful Azure cloud ecosystem available to the space community. Today, Microsoft has announced new partnerships and capabilities for Azure Space, including innovations such as SpaceEye, which allows users to ‘see’ through clouds from satellites.

At any given time, two-thirds of the Earth is covered in clouds, making space-based observation of Earth much more challenging. Built on Azure by Microsoft Research, SpaceEye is an AI-based system that generates daily cloud-free optical and multispectral imagery of Earth. SpaceEye uses the Synthetic Aperture Radar (SAR) instrument from the Sentinel-1 mission to produce baseline data, since radar isn’t affected by cloud cover, and then combines the radar data with historical optical data to generate an AI image that predicts what a scene may look like under the clouds. Microsoft states that ‘This can unlock significant use cases in agriculture, land-use monitoring and disaster response among others.’

An example of SpaceEye in use. Credit: Microsoft

Microsoft Azure isn’t stopping there. The platform is also being used to prepare and enhance geospatial data. With Microsoft’s Turing research, Azure Space has brought the fictional ‘enhance image’ feature regularly seen in television shows and movies into reality. Turing uses semantic super-resolution to increase the resolution of satellite imagery, making it look more like atmosphere-based aerial photography. This technology is being used on Bing Maps worldwide, covering more than 50 percent of user requests.

Microsoft has partnered with Airbus for Azure Space to expand Microsoft’s ‘mission to make Azure Space the platform and ecosystem of choice for the space community.’ Airbus imagery and elevation data are now available in Azure Maps. If you’re a customer of Azure Maps, you can now access a new global reference layer comprised of Airbus’s high-quality imagery and a homogenous elevation dataset that covers the entire globe.

With the Microsoft Research Team’s Turing process, satellite images can be enhanced to show much better detail. Credit: Microsoft

Azure Orbital uses virtualization to move connectivity functionality from expensive, inflexible radio hardware to a software ecosystem that can be deployed on general-purpose hardware. This makes connectivity more scalable and cost-efficient. Cost is not the only roadblock to the space community. There’s also an issue with a lack of standardization. Microsoft Azure is working with the Digital IF Interoperability (DIFI) consortium, of which Microsoft is a founding member, to create standards for the space ecosystem, including the new IEEE-ISTO Std 4900-2021: Digital IF Interoperability Standard for streaming data between digitizers and virtualized modems.

Azure Orbital promises to let users manage satellite data at a cloud scale, addressing an increasing need for low latency, high capacity ground infrastructure without forcing users to invest in expensive ground-based networks. ‘Microsoft is announcing the next milestone for Azure Orbital with the preview of a service that enables satellite operators to eliminate the technical, scheduling, and cost challenges of building a dedicated ground station network,’ Microsoft says. ‘The scale and scope of global ground stations through Azure Orbital and our partner network makes coverage more accessible for satellite providers at a lower cost, enabling high reliability and resiliency around the world ensuring support for any mission profile.’

The Azure Orbital preview includes support for Microsoft and KSATlite ground stations. Beginning this month, customers can use Azure Orbital APIs or the Azure Portal to communicate with satellites using Microsoft and KSAT antennas. KSAT CEO, Rolf Saktteboe, said, ‘Over the past year, we have worked hard here at KSAT to continue our collaboration with Azure Orbital integrating our worldwide satellite ground network with Microsoft to provide seamless, global support for transporting, processing, and storing space-based data. Being able to deliver satellite data and run resource-intensive computing such as machine learning techniques and other applications using cloud-based solutions will not only change the way we deliver our services but also how our customers will be able to utilize this information in the future. Through this partnership, our goal is to continue to provide our customers with the most technically advanced solutions for their missions.’

Microsoft has also announced it’s added ST Engineering iDirect, a trusted and widely-used satellite platform, to the Azure Orbital ground segment partners list. Microsoft writes, ‘iDirect and Microsoft are transforming satellite ground stations into a fully virtualized digital platform enabling Satellite operators to achieve the economies of scale and software efficiencies of cloud scale operation.’

In addition to partnering with Airbus and iDirect, Microsoft has also announced new partnerships with Esri (market leader in geographic information system software, location intelligence and mapping), Blackshark.ai (geospatial intelligence and 3D synthetic environments) and Orbital Insight (geospatial analytics software company).

Author:
This article comes from DP Review and can be read on the original site.

Related Posts