A linear approximation based method for noise-robust and illumination-invariant image change detection

Image change detection plays a very important role in real-time video surveillance systems. To deal with the illumination, a category of linear algebra based algorithms were designed in the literature. They have been proved to be effective for surveillance environment with lighting and shadowing. In practice. other than illumination, the detecting process is also influenced by the noises of cameras and reflections. In this paper, analysis is made systemically on the existing linear algebra detectors, showing their intrinsic weakness in case of noises. In order to get less sensitive to noises, a novel method is proposed based on the technique of linear approximation. Theoretical and experimental analysis both show its robustness and high performance for noisy image change detection.