How to Take a Picture of a Black Hole

The first-ever picture of a black hole was unveiled yesterday, generating a huge amount of interest and excitement across the world. But how exactly was this picture captured? Well, it definitely wasn’t as simple as “pointing and clicking” a giant camera.

One of the researchers who has become a poster child of the black hole photo project is MIT grad student Katie Bouman, who helped to develop the algorithm that turned raw telescope data into the black hole picture. A photo she posted on Facebook of herself watching the groundbreaking photo being reconstructed has gone wildly viral online over the past day:

Back in mid-2017, Bouman gave a TED Talk about her work, explaining how scientists were trying to image a black hole, and predicting that the first-ever photo would arrive in a couple of years.

“We may be seeing our first picture of a black hole in the next couple years,” Bouman stated. “Getting this first picture will come down to an international team of scientists, an Earth-sized telescope, and an algorithm that puts together the final picture.”

Capturing an image of a black hole requires an Earth-sized telescope due to the laws of diffraction, but building a physical one is pretty much impossible. But instead of blanketing the whole Earth with telescope mirrors, you could also use a much, much smaller number of telescopes spread out across the world and then extrapolate the data you collect — in other words, you can use algorithms to fill in the missing pieces.

Bouman’s algorithm, which she named CHIRP (Continuous High-resolution Image Reconstruction using Patch priors), is designed to do just that. It takes the data collected by telescopes around the world in the Event Horizon Telescope (EHT) project and figures out the most likely picture that fits the data.

By doing this, researchers have a computational telescope the size of Earth and can begin to accomplish imaging feats that were once considered impossible.

The black hole that the Event Horizon Telescope was formed to study is about the same size in the sky to us here on Earth as an orange on the surface of the Moon. Researchers at telescopes around the world froze light at exactly the same times over 9 days in April 2017 by synchronizing to atomic blocks, generating petabytes of data.

The massive amounts of data were then sent in for processing at a lab at MIT. But because of how much data it was, scientists couldn’t simply send it over the Internet — hard drives had to be physically shipped from all corners of the world before the data-crunching using Bouman’s algorithm could be completed.

Left: MIT computer scientist Katie Bouman w/stacks of hard drives of black hole image data.

Right: MIT computer scientist Margaret Hamilton w/the code she wrote that helped put a man on the moon.

(image credit @floragraham)#EHTblackhole #BlackHoleDay #BlackHole pic.twitter.com/Iv5PIc8IYd

— MIT CSAIL (@MIT_CSAIL) April 10, 2019