New imaging technique provides "drone's eye" view of Mars
University College London (UCL) has released images of the Martian surface with five times the resolution of anything previously sent back from Mars orbit. The images come courtesy of the new Super-Resolution Restoration (SRR) imaging technique developed by a UCL research team, which takes images from spacecraft orbiting Mars and stacks and matches them to create new, more detailed images of the Beagle 2 lander, ancient Martian lake beds, and the tracks of the NASA MER-A rover.
The fleet of orbiters now circling the Red Planet have added a bonanza of data to our knowledge and in four decades have mapped Mars with a detail that took centuries to do for the Earth. Unfortunately, the images they can send back is limited by the size of the telescopes that can be sent with the orbiters. According to UCL, these smaller optics combined with atmospheric interference and restricted bandwidth of Mars-Earth communications limits image resolution to about 25 cm (10 in).
To help overcome this, UCL's SRR takes existing images taken from orbit at different angles and stacks and matches them to produce images that can resolve objects as small as 5 cm (2 in) in diameter. In the recent study, the team took stacks of four to eight 25-cm images from the NASA HiRISE camera, which came back with high enough resolution to zoom in on specific objects like the failed Beagle 2 lander.
"Using novel machine vision methods, information from lower resolution images can be extracted to estimate the best possible true scene," says Yu Tao, Research Associate at UCL. "This technique has huge potential to improve our knowledge of a planet's surface from multiple remotely sensed images. In the future, we will be able to recreate rover-scale images anywhere on the surface of Mars and other planets from repeat image stacks."
UCL says that the technique was used on Mars because most features on the Martian surface remain stable for millions of years. As a result, the researchers were able to use images taken over a 10-year period and match them with, in the words of Jan-Peter Muller of the UCL Mullard Space Science Laboratory, "the equivalent of drone-eye vision anywhere on the surface of Mars where there are enough clear repeat pictures."
By contrast, Earth is a very poor candidate because it is so dynamic due to atmospheric turbulence that images can alter dramatically within seconds of one another.
According to UCL, the new images provide new evidence that the landing site of Beagle 2 has been correctly identified and the team hopes that SRR can find not only other failed landings, but also identify safe landing sites for future missions and provide rover-level images from orbit.
"As more pictures are collected, we will see increasing evidence of the kind we have only seen from the three successful rover missions to date," says Muller. "This will be a game-changer and the start of a new era in planetary exploration."
The research was published in Planetary and Space Science.