A joint effort between astrophysicists at the Institute for Advanced Study, Georgia Tech, and the NSF NOIRLab have produced a sharper image of M87, the supermassive black hole first visualized in 2019. A new machine-learning technique, named PRIMO, enhances the sharpness and fidelity of the radio interferometric image. The new algorithm offers a makeover to the “fuzzy, orange donut.”
![M87_PRIMO_compare_BW8_no_scale_text](https://bhpire.arizona.edu/wp-content/uploads/2023/04/M87_PRIMO_compare_BW8_no_scale_text-scaled.jpg)
M87 supermassive black hole originally imaged by the EHT collaborationin 2019 (left); and new image generated by the PRIMO algorithm using the same data set (right); Credit: Medeiros et al. 2023
![PRIMO_M87_fade_video](https://bhpire.arizona.edu/wp-content/uploads/2023/04/PRIMO_M87_fade_video.gif)
Animation fades from M87 black hole image,first producedby the EHT collaborationin 2019,to the new image generated by the PRIMO algorithm using the same data set; Credit: Medeiros et al. 2023