Multimedia Photogrammetry with non-planar Water Surfaces – Accuracy Analysis on Simulation Basis

If multimedia-photogrammetry is used for the generation of point clouds of submerged objects or of the water bottom, Snell’s law has to be considered. When the images are taken from air, image rays are refracted at the air-water interface. This results in the collinearity equations being no longer valid. Bundle block adjustment can still be solved by adding additional terms considering Snell’s law. Existing approaches usually assume that the water surface is flat. Refractive indices and water height can either be measured separately or included as unknowns in the adjustment. However, when the water surface is not flat due to the presence of waves, assuming a planar water surface leads to large geometric errors. This work will analyze the significance of those errors and propose a way of including water surface parameters as unknowns into the bundle block adjustment, both based on simulated data. The simulation reproduces multiple images taken simultaneously, e.g. from synchronized UAV cameras or from cameras on tripods.