A robust moving shadow detection algorithm based on semi-supervised hierarchical mixture of MLP-experts
Hamidreza Shayegh Boroujeni, Nasrollah Moghadam Charkari, Ali Jalilvand · 2011
Detection and elimination of the shadows of moving objects in video sequences have been one of the major challenges in tracking applications. Since moving shadows can't be removed from foreground by background subtraction methods, they lead to confusion and error in moving object tracking. In this paper, we propose a novel classification method based on hierarchical mixture of experts learning for detecting shadows from foreground. We propose Hierarchical Mixture of MLP Experts method (HMOE) with semi-supervised learning (SSP-HMOE) that uses a two level MOE system for shadow detection. The main superiority of this method is that it is almost robust and it works in all types of indoor and outdoor environments without any restrictions on the number of light sources, illumination conditions, surface orientations, object sizes, etc. The result of experiments in outdoor and indoor environments show the validity of the method in the improvement on the accuracy of both detection and discrimination rate for moving shadows in video sequences. The results of the experiments show the accuracy rate of 92% in average in different indoor and outdoor environmental conditions that is about 6% better than well-known similar methods.