Detection of Cast Shadows in Surveillance Applications
Søren G. Erbou, Helge Bjarne Dissing Sørensen, Bjarne Stage · 2005
Cast shadows from moving objects reduce the general ability of robust classi cation and tracking of these objects, in outdoor surveillance applications. A method for segmentation of cast shadows is proposed, combining statistical features with a new similarity feature, derived from a physics-based model. The new method is compared to a reference method, and found to improve performance signi cantly, based on a test set of real-world examples.