SIMBAD references

Ultra-Faint Dwarf Galaxies (UFDs) are among the lowest-mass and least-luminous galaxies known. Using algorithmic and visual inspection, they are first identified as galaxy candidates in ground-based imaging. UFDs at distances greater than ∼300 kpc are subsequently verified through space-based observations. However, the oversubscription of space-based telescopes, coupled with the difficulty of resolving faint systems at these distances, restricts the number of confirmed UFDs. Hence, the training data available to validate and refine existing search algorithms remains limited. We present a pipeline that combines existing tools to generate artificial UFDs that closely resemble known UFDs in ground-based images. We further extract photometric information with realistic observational uncertainties. This pipeline significantly increases the available training data with tens of thousands of artificial candidates, enabling advancement of UFD search algorithms and the discovery of UFDs beyond the current observational limits.