PREDICTING PHYSICAL PARAMETERS OF BLACK IMAGES USING GENERATED IMAGES
ABSTRACT
The Event Horizon Telescope (EHT) Collaboration were the first to release the first image of a black hole at the center of M87 Galaxy using Computer vision and Machine learning techniques. With the Very Long Baseline Inteferometry (VLBI) technique, the EHT group leveraged the highest possible resolution on the surface of the Earth. Based on this, further simulations were performed to get more black hole images. Using this simulated data, previous work proposes a new AI-assisted parameterization method that extracts important physical parameters from the high resolution images of M87 black hole. However, each of this simulation is computationally expensive and is time consuming. Hence, in this work we propose predicting the observing conditions choices to get par- ticular spin and magnetization values for image simulation or observation.
