Adaptive shadow estimator for removing shadow of moving object
Computer Vision and Image Understanding
Efficient ghost removal in motion detection with patch-corrected background differentiation
Optical Memory and Neural Networks
| Hi-index | 0.00 |
Chang detection is one of the most important research issues in the field of video processing. This paper presents an adaptive method of objects and shadows detection in video streams based on the HSI color space. Bi-models of background is set up via the minimum, the maximum and the largest interframe absolute difference of per static pixel, which are adaptively updated by synthesizing pixel level, object level and frame level method. In addition, motion region are built by three-level region chain structure to deal with the errors' detection results. Finally, we remove cast shadows using the generic properties of luminance, chrominance and gradient density in HSI color space. Experimental results and their evaluation verify the effectiveness and the robustness of the proposed approach.