Suppression of Cast Shadows for Improved Object Discrimination in Video Sequences
Ayoub K. Al-Hamadi, B. Michaelis · 2006
Firstly we apply a modified difference image-based approach for the extraction of moving objects in video sequences. Then we use post-processing algorithms and shadow detection filter for suppression of the remaining errors by the segmentation. The goal of the shadow detection algorithm is to prevent moving shadows being misclassified as moving objects, thus avoiding the merging of two or more objects into one and improving the accuracy of object description and localization. The efficiency of this approach is demonstrated on the basis of the analysis of disturbed real image sequences