Orbital Insight launches multi-class target detection algorithms for ships, aircrafts and vehicles

2021-12-14 23:03:33 By : Ms. Cathy Shen

Press release from: Orbital Insight Partners Release time: Tuesday, December 14, 2021

Orbital Insight, a leader in the field of geospatial intelligence, today announced the launch of multi-class object detection algorithms on its GO platform. Multi-class target detection uses advanced computer vision algorithms to identify and distinguish the types of ships, aircraft, and vehicles in satellite images. The most advanced multi-level ship detection algorithm will bring direct benefits to the security, supply chain and other critical needs of the defense and intelligence community.

Orbital Insight’s flagship GO platform is built specifically for the Department of Defense/intelligence community to provide insight in a cloud-agnostic, secure, and easy-to-deploy environment. GO will be deployed to government networks this year. GO combines information from the world's sensors (including satellites, aviation, AIS, and IoT devices) to analyze economic, social, and environmental trends on a large scale. The technology supports key activity-based intelligence by helping to discover trends hidden in billions of data points, such as the patterns of global ports, airports, and buildings. The platform is also a valuable tool for business sectors such as financial services and real estate to understand changing trends.

Each new multi-class target detection algorithm will help the military and intelligence communities to automatically monitor activities in thousands of areas of interest. Aircraft detection can detect and separate fighter jets, bombers, helicopters, commercial jets, etc.; ship detection can identify fleets of warships, cargo ships, fishing boats, etc.; vehicle detection can distinguish cars, trucks, rail cars, and upcoming mobile missile launchers and tanks .

"Knowing the movement of specific ships, aircraft, and railcars is critical to our national security," said Kevin O'Brien, CEO of Orbital Insight. "Computer vision can achieve this at scale and granular analysis, which humans cannot do by themselves. Through these multi-type object detection algorithms, our customers will have a clearer understanding of activities that are directly beneficial to us. The security posture."

Orbital Insight works with multiple strategic satellite and data partners to ensure that customers can get the latest and most accurate view of what is happening on and on the earth. For multi-class object detection, Orbital Insight uses Planet’s SkySat constellation and Airbus’s Pleiades to obtain clear, high-resolution global satellite images. Successful algorithms are the result of training models on hundreds of thousands of images and overcoming challenges such as clustered objects, weather obstacles, and low data set diversity.

"The combination of Planet's 50 cm SkySat data and Orbital Insight's advanced analysis allows users to gain a deeper understanding of the changing conditions on the ground," said Jim Thomason, Planet's vice president of imaging and analysis products. "We are eager to see the use cases that this hierarchical data set will unlock, because users can identify features more quickly and make decisions in the areas that are most important to them."

Like all Orbital Insight products, multi-type object detection was developed within an ethical framework that shapes the company's work and values ​​privacy.

To learn more about Orbital Insight, please visit www.orbitalinsight.com.

Orbital Insight is a geospatial software and analysis company that helps organizations understand what is happening on and on the earth. Customers including Unilever, Airbus, Royal Bank of Canada Capital Markets, the World Bank, and the U.S. Department of Defense use Orbital Insight's self-service analysis platform to make smarter business decisions, establish a sustainable supply chain, and improve national security. For more information, please visit www.orbitalinsight.com.

Mission North's orbital insight 

More press releases and status reports or headlines.

Please follow SpaceRef on Twitter and like us on Facebook.