Skip to main content

Machine learning used to sharpen the first image of a black hole

The world watched in delight when scientists revealed the first-ever image of a black hole in 2019, showing the huge black hole at the center of galaxy Messier 87. Now, that image has been refined and sharpened using machine learning techniques. The approach, called PRIMO or principal-component interferometric modeling, was developed by some of the same researchers that worked on the original Event Horizon Telescope project that took the photo of the black hole.

That image combined data from seven radio telescopes around the globe which worked together to form a virtual Earth-sized array. While that approach was amazingly effective at seeing such a distant object located 55 million light-years away, it did mean that there were some gaps in the original data. The new machine learning approach has been used to fill in those gaps, which allows for a more sharp and more precise final image.

A team of researchers, including an astronomer with NSF’s NOIRLab, has developed a new machine-learning technique to enhance the fidelity and sharpness of radio interferometry images. To demonstrate the power of their new approach, which is called PRIMO, the team created a new, high-fidelity version of the iconic Event Horizon Telescope's image of the supermassive black hole at the center of Messier 87, a giant elliptical galaxy located 55 million light-years from Earth. The image of the M87 supermassive black hole originally published by the EHT collaboration in 2019 (left); and a new image generated by the PRIMO algorithm using the same data set (right).
The image of the M87 supermassive black hole originally published by the Event Horizon Telescope collaboration in 2019 (left); and a new image generated by the PRIMO algorithm using the same data set (right). L. Medeiros (Institute for Advanced Study), D. Psaltis (Georgia Tech), T. Lauer (NSF’s NOIRLab), and F. Ozel (Georgia Tech)

“With our new machine-learning technique, PRIMO, we were able to achieve the maximum resolution of the current array,” said lead author of the research, Lia Medeiros of the Institute for Advanced Study, in a statement. “Since we cannot study black holes up close, the detail in an image plays a critical role in our ability to understand its behavior. The width of the ring in the image is now smaller by about a factor of two, which will be a powerful constraint for our theoretical models and tests of gravity.”

PRIMO was trained using tens of thousands of example images which were created from simulations of gas accreting onto a black hole. By analyzing the pictures that resulted from these simulations for patterns, PRIMO was able to refine the data for the EHT image. The plan is that the same technique can be used for future observations from the EHT collaboration as well.

“PRIMO is a new approach to the difficult task of constructing images from EHT observations,” said another of the researchers, Tod Lauer of NSF’s NOIRLab. “It provides a way to compensate for the missing information about the object being observed, which is required to generate the image that would have been seen using a single gigantic radio telescope the size of the Earth.”

In 2022, the EHT collaboration followed up its image of the black hole in M87 with a stunning image of the black hole at the heart of the Milky Way, so that image could be the next target for sharpening using this technique.

“The 2019 image was just the beginning,” said Medeiros. “If a picture is worth a thousand words, the data underlying that image have many more stories to tell. PRIMO will continue to be a critical tool in extracting such insights.”

The research is published in The Astrophysical Journal Letters.

Editors' Recommendations

Georgina Torbet
Georgina is the Digital Trends space writer, covering human space exploration, planetary science, and cosmology. She…
Supermassive black hole spews out jet of matter in first-of-its-kind image
Scientists observing the compact radio core of M87 have discovered new details about the galaxy’s supermassive black hole. In this artist’s conception, the black hole’s massive jet is seen rising up from the centre of the black hole. The observations on which this illustration is based represent the first time that the jet and the black hole shadow have been imaged together, giving scientists new insights into how black holes can launch these powerful jets.

As well as pulling in anything which comes to close to them, black holes can occasionally expel matter at very high speeds. When clouds of dust and gas approach the event horizon of a black hole, some of it will fall inward, but some can be redirected outward in highly energetic bursts, resulting in dramatic jets of matter that shoot out at speeds approaching the speed of light. The jets can spread for thousands of light-years, with one jet emerging from each of the black hole's poles in a phenomenon thought to be related to the black hole's spin.

Scientists observing the compact radio core of M87 have discovered new details about the galaxy’s supermassive black hole. In this artist’s conception, the black hole’s massive jet of matter is seen rising up from the center of the black hole. The observations on which this illustration is based represent the first time that the jet and the black hole shadow have been imaged together, giving scientists new insights into how black holes can launch these powerful jets. S. Dagnello (NRAO/AUI/NSF)

Read more
Previously unknown exoplanet discovered using machine learning
exoplanet discovered machine learning image png

When it comes to discovering new astronomical bodies, sometimes humans are irreplaceable thanks to their skills in pattern detection. But in other cases, computers can spot things that aren't visible to humans -- including a recent instance where an exoplanet was discovered using machine learning.

The exoplanet was discovered by University of Georgia researchers within a protoplanetary disk called HD 142666. A protoplanetary disk is a rotating disk of gas that swirls around young stars, and from which planets are formed. Planets are formed within these disks as matter clumps together until it eventually has enough gravity to pull more material in. The researchers looked at a previous set of observations of a whole set of protoplanetary disks, and used a machine learning model to search for exoplanets that might have been missed the first time around. They identified one disk where a planet was likely to be, based on the unusual way that gas moved around within the disk.

Read more
Unique black hole is trailed by 200,000 light-year-long tail of stars
This is an artist's impression of a runaway supermassive black hole that was ejected from its host galaxy as a result of a tussle between it and two other black holes. As the black hole plows through intergalactic space it compresses tenuous gas in front to it. This precipitates the birth of hot blue stars. This illustration is based on Hubble Space Telescope observations of a 200,000-light-year-long contrail of stars behind an escaping black hole.

Black holes might have a reputation as terrifying monsters, devouring all they come into contact with -- but they can be a force of creation too, feeding the formation of new stars. Researchers using data from the Hubble Space Telescope recently spotted an unexpectedly huge trail of stars forming in the wake of a rogue black hole.

While most very large black holes, called supermassive black holes, sit at the center of galaxies, occasionally these enormous beasts can be found wandering alone in the depths of space. That's the case with the recently discovered black hole with the mass of 20 million suns, which is streaking through the sky at tremendous speed. This likely began with two galaxies merging, each with its own supermassive black hole, which formed a binary system. Then a third galaxy got too close, and in the chaos of a three-way merger one of the black holes was kicked out and sent zipping off into space -- so fast that if it were in our solar system, it would travel from the Earth to the moon in 14 minutes.

Read more