Behind the algorithm capturing the black hole images

Since the start of this week, capturing images of an actual live black hole has been making the rounds on social and mainstream media. The team responsible for creating the algorithm and imaging techniques to capture the image has been working on it since 2016, with Katie Bouman as the project head.

Katie Bouman is a computer scientist who has been leading the project for developing the CHIRP(continuous High-resolution Image Reconstruction using Patch priors) algorithm.

The algorithm is responsible for creating images of the black hole that have been gaining a lot of traction on mainstream media as well as social media.

The algorithm uses data from the eight radio telescopes, spread across 4 continents to capture the image, who fall under the Event Horizon Telescope project. The project team has been dedicated to collecting the images of the celestial body and enhance its quality for viewing.

Katie was still a graduate of MIT when she started working on the algorithm. She is currently working on the Event Horizon Telescope, with a post-doctorate. According to her website, she will be joining Caltech’s computing and mathematical department as an assistant professor.

MIT made an announcement in 2016 that it will start working on the CHIRP algorithm. The core function of the project was to turn “to turn the entire planet into a large radio telescope dish.” The project team gathered researchers and team members from three places: MIT’s Computer Science and Artificial Intelligence Laboratory, the Harvard-Smithsonian Center for Astrophysics, and the MIT Haystack Observatory.

Image result for black hole picture

Hiccups along the way

The margin of error between the decoding of the astronomical signals by the radio telescopes posed quite a problem for the team regarding the calculations for getting an accurate picture. According to MIT, Katie came up with a solution:

“Bouman adopted a clever algebraic solution to this problem: If the measurements from three telescopes are multiplied, the extra delays caused by atmospheric noise cancel each other out. This does mean that each new measurement requires data from three telescopes, not just two, but the increase in precision makes up for the loss of information.”

The algorithm reconstructs and refines the image and has practical applications in any imaging system using radio interferometry. This is how the team achieved the objective of getting clear and refined images for a black hole.

The team’s data collection was quite large that the data had to be transferred to the MIT Haystack Observatory’ servers and hard drives.

Further details on the development of the algorithm by Bouman can be found below:

Let us know, what you think about images of the celestial body, in the comment section below.


Add comment