There is no doubt that artificial intelligence has facilitated the execution of processes in different areas, including space, where NASA has now managed to detect the presence of craters on the surface of Mars with the help of this technology.
In this sense, artificial intelligence was able to achieve the detection of craters thanks to the scanning of the photographs corresponding to the Mars Reconnaissance Orbiter, launched in 2005, to analyze the history of water on the red planet.
Scientists usually track these craters through a scanning process carried out with their own eyes on images captured by NASA's Mars Reconnaissance Orbiter Context Chamber.
These images represent an area of hundreds of kilometers at a time taken at low resolution. Each takes about 40 minutes to be scanned by the researcher.
That's why, to accelerate this process and foster a greater number of findings, scientists and AI researchers at NASA's Jet Propulsion Laboratory (JPL) in Southern California were tasked with creating a tool to which they attributed the name of Automated Fresh Impact Crater Classifier.
Initially, some 6,830 context camera images were used to train the classifier, which was then nurtured with 112,000 images representing the complete repository of this source. This significantly reduced the image analysis time from 40 minutes to 5 seconds, although the image still required human intervention to validate its work.
In this regard, Kiri Wagstaff, a JPL computer scientist, said:
«AI cannot do the kind of expert analysis that a scientist can but tools like this new algorithm can be your assistants. This paves the way for the exciting symbiosis of human and AI researchers working together to accelerate scientific discovery»
0 Comments