(ORDO NEWS) — In 2019, astronomers provided the first- ever image of a black hole located at the center of the M87 galaxy in the nearby Virgo cluster of galaxies.
Now, using a new machine learning technique, we have a clearer and more accurate representation of observations of the M87* black hole.
The astronomers involved in the study note that they did not use artificial intelligence to fill in the gaps, as some smartphones do to improve the imaging capabilities of their cameras.
Instead, they took raw data from the Event Horizon Telescope (EHT) collaboration and processed it using the vocabulary-based PRIMO program, where AI is taught “rules” by showing it thousands of examples.
Applying PRIMO to EHT M87* data allowed us to analyze 30,000 high-fidelity simulated event horizon images to find common patterns in the images.
Mixing the results gave a clearer and yet very accurate representation of the EHT observations, which also managed to estimate some of the missing data from the original image.
Their goal was ambitious: to depict the shadow of a supermassive black hole’s event horizon, which they did.
Prior to this first image, the only way to look at a black hole was to indirectly observe its effects, not the object itself.
Now, thanks to a new image processed with the help of AI, the dark central region of the object and the narrow bright ring of radiation from hot gas falling into the black hole are better visible.
This image will help scientists study the characteristics of the black hole and its surroundings in more detail.
Also, the new image will help confirm the theory that black holes in the center of galaxies are sources of powerful radiation and play an important role in the development of galaxies.
In addition, the development of new AI algorithms for processing cosmic data could lead to further discoveries and improve our understanding of the universe.
Contact us: [email protected]