Variational Image Feature Extraction for the Event Horizon Telescope

Imaging algorithms form powerful analysis tools for VLBI data analysis. However, these tools cannot measure certain image features (e.g., ring diameter) by their non-parametric nature. This is unfortunate since these image features are often related to astrophysically relevant quantities such as black hole mass. This paper details a new general image feature extraction technique that applies to a wide variety of VLBI image reconstructions called variational image domain analysis. Unlike previous tools, variational image domain analysis can be applied to any image reconstruction regardless of its structure. To demonstrate its flexibility, we analyze thousands of reconstructions from previous EHT synthetic datasets and recover image features such as diameter, orientation, and asymmetry. By measuring these features, VIDA can help extract astrophysically relevant quantities such as the mass and orientation of M 87.