This work demonstrates the capacity of a ballistic imaging instrument to suppress diffuse photons and improve image contrast, making it possible to view fluid structure in a spray where a fog of droplets occludes the near field. Analysis of the system is performed by means of a numerical system model. The model simulates light propagation and scattering in the measurement volume using a Monte Carlo based solution to the radiative transfer equation, and includes treatment of the full system optics using a custom ray-tracing code. Simulation results for the validation case where source light illuminates a test chart inside a turbid (optical depth OD = 14) solution of monodisperse polystyrene spheres (d = 0.7 mu m) show good agreement with experimental images. The model is further applied by replacing the solution of polystyrene spheres with a spraylike scattering medium. Here, we investigate the temporal characteristics of an ultrashort (100 fs) laser signal crossing a volume containing a polydisperse distribution of fuel droplets with a representative Sauter mean diameter, D-32 = 23 mu m. These quantitative predictions allow the effectiveness of both the spatial and temporal filtering of the ballistic imaging instrument to be estimated. Results from the model demonstrate that the spatial filtering and time gating effects of the ballistic imaging system significantly improve image contrast, revealing information that is not available with conventional imaging techniques.