Research Corporation Backs a Radical New Approach to Astronomical Imaging


A 12-megapixel digital camera can capture an image composed of roughly 12 million pieces of information and is all most of us need to take stunning vacation photos or family portraits suitable for framing. However, if you’re an astronomer taking detailed images of celestial bodies that are thousands to millions of light years away, it’s nowhere near good enough. But thanks to a Cottrell SEED Award from the Research Corporation for Scientific Advancement, UVA astronomer Steven Majewski may be one step closer to developing a technology that makes taking vivid panoramic snapshots of distant stars and galaxies something more than just the stuff of science fiction.  

Some of the largest telescopes used today can measure up to 10 meters in diameter, and there are plans to build them up to three times as large; however, capturing all the light that enters a telescope of that size and transforming it into a digital image would require a sensor approximately a thousand times larger than the ones found in the best cameras on the market. Astronomical cameras are being developed that can capture images on the scale of roughly a billion pixels – or a gigapixel – but the cost of building them could add up to hundreds of millions of dollars. Even for astronomers who have that kind of money to spend, there’s one big drawback: much of that additional capacity is wasted.

“When you take a picture of the sky, most of the picture is black. It’s dark sky, but what we’re interested in imaging are, of course, the things that have light,” said Majewski, a professor of astronomy with UVA’s College and Graduate School of Arts & Sciences.

The solution, Majewski believes, is not building cameras with more pixels but building ones that make better use of the pixels they have.

“My idea is: let’s think about a new paradigm,” Majewski said. “If pixels are precious and cameras are expensive to build, why don’t we just put the pixels where we need them and ignore the places where we don’t.”

Astronomer Steven Majewski has won a SEED Award to build a new kind of camera that could transform how we see the universe. (Sanjay Suchak, University Communications)

The camera that Majewski envisions – and wants to test with a prototype – is something he calls the Distributed Field Fiber Imaging Testbed, or DIFFIT. It uses flexible strands of fiber optic cable to allow astronomers to select only those parts of a telescope’s entire field of view that are of interest and delivers them directly to the camera.  In effect, such a device would capture smaller, cutout images in greater detail. It would also use pixels more economically, requiring a much smaller sensor – an idea that’s not entirely unprecedented.

“Astronomers have been using fiber optics like this for a very long time to collect all of the light from one source and put it into, say, a spectrograph where we can see the spectrum of that light source, but they haven’t done it for imaging, which is a tougher problem, because it requires preserving the spatial information,” Majewski said.  “On the other hand, the field of internal medicine routinely uses fiber optic endoscopes to collect images in hard-to-reach places and transport them somewhere more convenient. What I’m proposing is to build an astronomical camera with many endoscopic probes sampling only the interesting parts of a telescope field of view and delivering and repackaging those postage stamp images into a compact format.”

A New Look at the Universe

While the DIFFIT technology could be applied to many astronomical problems in the near term,  Majewski envisions applying it to better understanding the architecture and behavior of star systems that contain multiple stars, like binary stars that orbit around each other, and as a tool in efforts to discover and characterize planets around other stars. The vast majority of these “exoplanets” have been found through a process that involves observing stars that dim temporarily when planetary bodies pass in front of them. Typically, satellite-based telescopes are used to make observations of that kind because they are immune to the unrelated and much larger “twinkling” effect caused by air currents from heat convection, winds and the jet stream in the earth’s atmosphere. But because orbital telescopes have so much of space to scan in a limited amount of mission time, they’ve been built to have  “fat pixels,” or single pixels that contain multiple objects. And with resolution that poor, drawing absolute conclusions from those images can be extremely challenging. Follow-up observations and analysis from earthbound telescopes are needed to check the exoplanet candidates from these space missions.

Over the last few decades, a technology known as speckle imaging has allowed astronomers using ground-based telescopes to recover very detailed images of stars and star systems by combining many high-speed images to create a composite that is enhanced by a computer algorithm developed to take out the visual distortions caused by the earth’s atmosphere. It’s an effective solution but currently not a perfect one because it relies on making comparisons between interesting, complex sources and one or more simpler, “reference” sources, all ideally observed at the same time through the same atmospheric distortions. But traditional speckle imaging cameras are only able to focus on one small patch of sky at a time, so they normally cannot achieve their maximum resolution. Majewski’s DIFFIT technology, however, would give astronomers the ability to observe not only many reference stars but numerous interesting targets as well, all simultaneously over a wide area of sky.  The result is the ability to gather very high-resolution images of scientifically interesting star systems much more efficiently than ever before.

But the first step is proving that DIFFIT works. Majewski, who is principal investigator of the APOGEE project, is no stranger to using cutting-edge instrumentation to advance our knowledge about the universe.  The APOGEE team pioneered several breakthrough technologies to build a pair of fiber-optic fed, infrared spectrographs to determine the chemical compositions and the motions of over half a million Milky Way stars that are difficult or impossible to observe with conventional astronomical instruments. Nevertheless, Majewski has spent several years trying to find funding to build the DIFFIT prototype.

“It’s a radical concept that I’m proposing,” Majewski admitted. “One that hasn’t been tried before, and getting funding for risky things is very hard.”

When the larger funding agencies seemed reluctant to invest in a concept that might not work, Majewski turned to the Research Corporation for Scientific Advancement, a nonprofit that funds basic research in the physical sciences and that created the Cottrell Scholars program, which recognizes outstanding teacher-scholars who are involved in innovative research and are academic leaders in their discipline. A Cottrell Scholar since 1998, Majewski was eligible for the SEED (Singular Exceptional Endeavors of Discovery) Award, which grants $50,000 to researchers with creative research projects that it characterizes as “high risk/high reward.” It’s not necessarily an impressive amount of money, but it’s enough to help Majewski and his students prove that the DIFFIT concept is sound.

“When funding agencies have very restricted budgets, as they do right now, they tend to become more conservative and want to fund the sure thing,” Majewski said. “I really like that the Research Corporation recognizes this fact and created the SEED Award, which allows more established scientists to try out ideas that are a change of direction or maybe a little novel.

And, he added, “The payoff could be great. If this concept proves its worth, they’ve enabled a transformative technology.”

“Steve has had a good track record when it comes to thinking outside the box and bucking conventional wisdom,” says Craig Sarazin, chairman of the Department of Astronomy. “When he proposed the APOGEE project 15 years ago, some astronomers and their institutions would not join the project, saying it couldn’t be done. But when the APOGEE team delivered one of the most successful Milky Way surveys ever, the doubters and their institutions suddenly wanted in.”

“Steve’s new DIFFIT innovation may not work,” Sarazin added, “but if it does, it will transform astronomy.”