We report on imaging subsurface grain microstructure using picosecond ultrasonics. This approach relies on elastic anisotropy of crystalline materials where ultrasonic velocity depends on propagation direction relative to the crystal axes. Picosecond duration ultrasonic pulses are generated and detected using ultrashort light pulses. In materials that are transparent or semitransparent to the probe wavelength, the probe monitors gigahertz frequency Brillouin oscillations. The frequency of these oscillations is related to the ultrasonic velocity and the optical index of refraction. Ultrasonic waves propagating across a grain boundary experience a change in velocity due to a change in crystallographic orientation relative to the ultrasonic propagation direction. This change in velocity is manifested as a change in the Brillouin oscillation frequency. Using the ultrasonic propagation velocity, the depth of the interface can be determined from the location in time of the transition in oscillation frequency. A subsurface image of the grain boundary is obtained by scanning the beam along the surface. We demonstrate this subsurface imaging capability using a polycrystalline UO2 sample. Cross section liftout analysis of the grain boundary using electron microscopy was used to verify our imaging results.