A Scale-dependent Morphological Approach to Motion Segmentation

Accurate motion segmentation is the foundation of several high-level applications, such as surveillance and traffic monitoring systems. Here, we use a simple toggle operator, defined on a scaled morphological framework, to process the image frames in such a way that moving objects can be easily detected. The operator is able to cope with cast shadows and illumination changes, it is robust to false positives and sufficiently fast enough to be used in real-time applications. Experimental results show that the proposed approach outperforms existing methods besides having low computational cost.