#UniversityofToronto, #Ontario, #Canada, #MohamadAli-Dib, #NASA #algorithms, #lunarcraters, #Ari Silburt, #AI, #Artificialintelligence #NaturalSciencesandEngineeringResearchCouncilofCanada, #NSERC, #convolutionalneuralnetwork, #KristenMenou,
Toronto, Mar 14 (Canadian-Media): Researchers at the University of Toronto (U of T), Ontario have used a new technique based on the technology behind self-driving cars to measure the size and location of crater impacts on the moon, media reports said.
“When it comes to counting craters on the moon, it’s a pretty archaic method. Basically we need to manually look at an image, locate and count the craters and then calculate how large they are based off the size of the image. Here we’ve developed a technique from artificial intelligence that can automate this entire process that saves significant time and effort, ”Mohamad Ali-Dib, a postdoctoral researcher at the Centre for Planetary Sciences (CPS) at U of T Scarborough was reported to state.
Detailed view of the back side of moon in the vicinity of Crater No. 308 taken during the Apollo 11 mission (photo courtesy of NASA)
The research is currently under review in the journal Icarus and had which reportedly received funding from the Natural Sciences and Engineering Research Council of Canada (NSERC).
Desired results could not reportedly be got earlier by the development of algorithms to identify and count lunar craters.
“It’s the first time we have an algorithm that can detect craters really well, for not only parts of the moon, but also areas of Mercury,” Ali-Dib, who developed the technique along with alumnus Ari Silburt, postdoctoral researcher Chenchong Charles Zhu, and a group of researchers at CPS and the Canadian Institute for Theoretical Astrophysics (CITA) were reported to state.
Mohamad Ali-Dib: Courtesy of Ken Jones
The determination of its accuracy was done by first training the neural network by the researchers on a large data set covering two-thirds of the moon.
These trained network were then reportedly tested on the remaining third of the moon.
Using this strategy it was possible for the researchers to identify about 6,000 previously unidentified craters on the moon, twice as many craters as traditional manual counting.
The technique has reportedly been successfully used for computer vision to power robots and even self-driving cars and is based on a class of machine learning algorithms known as convolutional neural network.
The data had been taken by the algorithms from elevation maps gathered from orbiting satellites.
It was only after a series of workshops held at U of T Scarborough that the researchers were able to develop this technique co-organized by Associate Professor Kristen Menou.
This fact had reportedly thrown more light on how specific scientific problems could be tackled both by machine learning and deep learning.
“Tens of thousands of unidentified small craters are on the moon, and it’s unrealistic for humans to efficiently characterize them all by eye,” Ari Silburt, a former graduate student in U of T’s department of astronomy and astrophysics and now a postdoc at Penn State University was reported to stae.
“There’s real potential for machines to help identify these small craters and reveal undiscovered clues about the formation of our solar system.”
Knowledge of the size and location of craters on bodies like the moon reportedly enables the researchers to learn the history of our solar system and to better understand the distribution of material and the physics that occurred in the early stages of our solar system, said Ali-Dib.
The count of small craters further advances the knowledge of ages of large craters as well as to find more craters.
“For this technique to work you need an airless body like the moon or Mercury, bodies where there’s little erosion taking place,” adds Ali-Dib.
Ali-Dib and his team also reportedly intend to test this technique on other solar system bodies like Mars, Ceres and the icy moons of Jupiter and Saturn.
(Reporting by Asha Bajaj)