A group of astronomers and computer scientists at the University of Hertfordshire have taught a machine to ‘see’ astronomical images.
The technique uses unsupervised machine learning – a form of artificial intelligence – that allows galaxies to be classified in real time at high speed, something previously done by thousands of human volunteers in projects like Galaxy Zoo.
The team demonstrated its algorithm using data from the Hubble Space Telescope ‘Frontier Fields’, including images of distant clusters of galaxies that contain several different types of galaxy, which the machine is now able to identify.
Masters student Alex Hocking, who led the work, said: “The important thing about our algorithm is that we have not told the machine what to look for in the images, but instead taught it how to see.”
His supervisor Dr James Geach said: “A human looking at these images can intuitively pick out and instinctively classify different types of object without being given any additional information. We have taught a machine to do the same thing.
“Our aim is to deploy this tool on the next generation of giant imaging surveys where no human, or even group of humans, could closely inspect every piece of data”.
“This algorithm has a huge number of applications far beyond astronomy and investigating these applications will be our next step,” Geach added.
The scientists are now looking for collaborators, making good use of the technique in applications like medicine, where it could help doctors to spot tumours, and in security, to find suspicious items in airport scans, for example.
In 2007, millions of photos were uploaded to a website called Galaxy Zoo and volunteers were sought to look through the photos and identify the different shapes of the galaxies so that scientists could learn more about them. Over an 18-month span, about 150,000 volunteers successfully classified 50 million galaxies, boosting scientific research.
The paper will be presented on Wednesday at the National Astronomy Meeting in Llandudno, Wales.