Astronomers teach a machine to analyse galaxy images

July 8, 2015, Royal Astronomical Society
Hubble Space Telescope image of the cluster of galaxies MACS0416.1-2403, one of the Hubble ‘Frontier Fields’. Bright yellow ‘elliptical’ galaxies can be seen, surrounded by numerous blue spiral and amorphous (star-forming) galaxies. Gravitational arcs can also be seen. This image forms the test data that the machine learning algorithm is applied to, having not previously ‘seen’ the image. Credit: NASA / ESA / J. Geach / A. Hocking

A team of astronomers and computer scientists at the University of Hertfordshire have taught a machine to 'see' astronomical images. The technique, which uses a form of artificial intelligence called unsupervised machine learning, allows galaxies to be automatically classified at high speed, something previously done by thousands of human volunteers in projects like Galaxy Zoo. Masters student Alex Hocking led the new work and presented it for the first time in a paper today (July 8) at the National Astronomy Meeting at Venue Cymru, Llandudno, Wales.

The team have demonstrated their algorithm using data from the Hubble Space Telescope 'Frontier Fields': exquisite images of distant clusters of galaxies that contain several different types of galaxy.

Mr Hocking, who led the new work, commented: "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 and fellow team member Dr James Geach added: "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. But this algorithm has a huge number of applications far beyond astronomy, and investigating these applications will be our next step," concludes Geach.

The scientists are now looking for collaborators, making good use of the technique in applications like medicine, where it could for example help doctors to spot tumours, and in security, to find suspicious items in airport scans.

Visualisation of the neural network representing the ‘brain’ of the machine learning algorithm. The intersections of lines are called nodes, and these represent a map of the input data. Nodes that are closer to each other represent similar features within the data. Fainter lines show how the network has evolved over time as the algorithm processes more data. Credit: J. Geach / A. Hocking

A zoom-in of part of the network described above. Credit: J. Geach / A. Hocking
Image showing the MACS0416.1-2403 cluster, highlighting parts of the image that the algorithm has identified as ‘star-forming’ galaxies. Credit: NASA / ESA / J. Geach / A. Hocking
Image showing the MACS0416.1-2403 cluster, highlighting parts of the image that the algorithm has identified as ‘elliptical’ galaxies. Credit: NASA / ESA / J. Geach / A. Hocking

Explore further: We're not alone—but the universe may be less crowded than we think

Related Stories

Public maps out an A to Z of galaxies

September 10, 2012

Volunteers participating in the Galaxy Zoo project have been helping scientists gain new insights by classifying galaxies seen in hundreds of thousands of telescope images as spiral or elliptical. Along the way they've also ...

When did galaxies settle down?

October 30, 2014

Astronomers have long sought to understand exactly how the universe evolved from its earliest history to the cosmos we see around us in the present day. In particular, the way that galaxies form and develop is still a matter ...

Strange galaxy perplexes astronomers

December 2, 2014

With the help of citizen scientists, a team of astronomers has found an important new example of a very rare type of galaxy that may yield valuable insight on how galaxies developed in the early Universe. The new discovery ...

Looking back to the cradle of our universe

February 10, 2014

(Phys.org) —NASA's Spitzer and Hubble Space Telescopes have spotted what might be one of the most distant galaxies known, harkening back to a time when our universe was only about 650 million years old (our universe is ...

Astronomers unveil the farthest galaxy

May 5, 2015

An international team of astronomers led by Yale University and the University of California-Santa Cruz have pushed back the cosmic frontier of galaxy exploration to a time when the universe was only 5% of its present age.

Recommended for you

Hubble paints picture of the evolving universe

August 16, 2018

Astronomers using the ultraviolet vision of NASA's Hubble Space Telescope have captured one of the largest panoramic views of the fire and fury of star birth in the distant universe. The field features approximately 15,000 ...

Sprawling galaxy cluster found hiding in plain sight

August 16, 2018

MIT scientists have uncovered a sprawling new galaxy cluster hiding in plain sight. The cluster, which sits a mere 2.4 billion light years from Earth, is made up of hundreds of individual galaxies and surrounds an extremely ...

Unusual doughnut-shaped jet observed in the galaxy NGC 6109

August 15, 2018

Astronomers from the University of Bristol, U.K., have uncovered an unusual doughnut-shaped jet in the radio galaxy NGC 6109. It is the first time that such a jet morphology has been observed in a low-power radio galaxy. ...

Iron and titanium in the atmosphere of an exoplanet

August 15, 2018

Exoplanets, planets in other solar systems, can orbit very close to their host stars. When the host star is much hotter than the sun, the exoplanet becomes as hot as a star. The hottest "ultra-hot" planet was discovered last ...

0 comments

Please sign in to add a comment. Registration is free, and takes less than a minute. Read more

Click here to reset your password.
Sign in to get notified via email when new comments are made.