Abstract
Denoising of Time-of-Flight (ToF) range data is an important task prior to further data processing. Existing techniques commonly work on a post processing level. This paper presents a novel approach for improving data quality on the image acquisition level by automatically determining the best integration time for arbitrary scenes. Our approach works on a per-pixel basis and uses knowledge gained from an extensive analysis of the underlying inherent sensor behavior regarding intensity, amplitude and distance error to reduce the overall error, to prevent oversaturation and to minimize the adaption time. It also works well in presence of various reflectivities and quick changes in the scene. This represents a significant improvement over previous methods.