Back in 2019 scientists made history after they released the first-ever image of a black hole, which cannot be seen with the naked eye. The team was able to photograph the dazzling silhouette of extremely hot gas and plasma swirling around the black hole at the center of the Messier 87 (M87) galaxy using the Event Horizon Telescope (EHT).
However, in a new development, researchers have used artificial intelligence to give that cosmic beauty shot a touch-up. The image of the supermassive black hole at the heart of the Galaxy Messier 87 was boosted to high fidelity by a machine learning program trained on black hole models.
How AI Modify the First-Ever Black Hole Image
Scientists used machine learning to give old photos a small makeover. The black hole appears as a somewhat thinner orange doughnut in the new image, which also reveals more of the blackness of the object itself and a clearer outline of the brilliant, superheated gas that surrounds it.
Principal-component interferometric modeling (PRIMO), a machine-learning approach, was utilized by the researchers to produce this new black hole image. On Thursday, they submitted a manuscript detailing the procedure to The Astrophysical Journal of Letters.
“With our new machine learning technique, PRIMO, we were able to achieve the maximum resolution of the current array,” Lia Medeiros, an astrophysicist at the Institute for Advanced Study and lead author of the paper, said in a statement. She added that the image allows researchers to gain a better understanding of black hole behaviors since “we cannot study black holes up close.”
The beauty of a technology like PRIMO is that it fills in the gaps left by our existing ways of seeing black holes utilizing the EHT, which necessitates a network of seven telescopes all over the world. There are gaps in the data because these telescopes can only observe a limited area of the black hole; PRIMO fills these gaps.
More than 30,000 simulated photos of black holes gathering gas are used to train the model. As a result, PRIMO was able to make reasonable assumptions about what M87 appears to be when viewed more clearly.
“We are using physics to fill in regions of missing data in a way that has never been done before by using machine learning,” Medeiros explained.
“This could have important implications for interferometry, which plays a role in fields from exoplanets to medicine.”
