Computer vision firm Metaspectral, along with remote sensing data analysis firm, HySpeed, are to design a hyperspectral imaging Earth observation payload for the International Space Station.
The system, known as ‘Onboard Programmable Technology for Image Classification and Analysis’ (OPTICA) and will enable real-time compression, streaming, and analysis of hyperspectral data from low Earth orbit.
OPTICA is scheduled for launch in early 2023 on the SpaceX CRS-27 mission, with a six-month deployment on the ISS following soon after. The mission is sponsored by the ISS National Laboratory, which works in collaboration with Nasa to use the orbiting laboratory on the ISS to its fullest research and technology development potential.
Metaspectral will be providing the hardware and software for the payload, with HySpeed Computing being responsible for creating the necessary data processing pipeline and analysis tools.
The system will use data compression and machine learning to overcome some of the bandwidth constraints when downlinking hyperspectral image data to Earth for analysis. Francis Doumet, Metaspectral co-founder and CEO, said the ‘project will demonstrate our platform’s ability to produce actionable insights [from the data] within 15 minutes or less’.
Doumet added: ‘Great strides have been made with optical inter-satellite links in increasing the reliability of satellite communications, and our technology complements this by bringing in advanced data compression and streaming for hyperspectral data from orbit.’
- The hyperspectral view from space - Abigail Williams speaks to scientists tracking marine plastic using satellite spectral imagery
Hyperspectral imagery captured by satellites contains data from across the electromagnetic spectrum which, when analysed with artificial intelligence, can be used to identify events on the Earth such as forest fires, methane leaks, and oil spills. It can also provide time-sensitive data to intelligence, surveillance, or reconnaissance missions.
‘OPTICA represents the next evolution in space- and ground-based image processing,’ said Dr James Goodman, CEO of HySpeed Computing. ‘As the volume, variety, and velocity of Earth observation data continue to increase, developing efficiencies in data processing and information delivery will be paramount throughout the remote sensing industry. OPTICA addresses this need by demonstrating the ability to acquire rapidly, process, and analyse imagery from a high-data-volume hyperspectral sensor.'
Metaspectral and HySpeed Computing have selected Nanoracks, a commercial space services provider, to serve as the implementation partner for the project.