Background models for tracking objects in water

This paper presents a novel background analysis technique to enable robust tracking of objects in water- based scenarios. Current pixel-wise statistical background models support automatic change detection in many outdoor situations, but are limited to background changes which can be modeled via a set of per-pixel spatially uncorrelated processes. In water-based scenarios, waves caused by wind or by moving vessels (wakes) form highly correlated moving patterns that confuse traditional background analysis models. In this work we introduce a framework that explicitly models this type of background variation. The framework combines the output of a statistical background model with localized optical flow analysis to produce two motion maps. In the final stage we apply object-level fusion to filter out moving regions that are most likely caused by wave clutter. A tracking algorithm can now handle the resulting set of objects.