An Adapting to Light Change Pixel Layer Based Background Model for Moving Objects Detection in a Dynamic Scene

A novel background model based on pixel layer for moving objects detection in a dynamic scene is one of the common methods in motion detection. This paper proposes an adapting to light change pixel layer based background model for moving objects detection and this improved model can deal with the problem of illumination change and complexity of background updating. First, a new model parameter was selected, a fixed learning rate was used and the adaptive factor to update the variance was added which are helpful to adapt to illumination change rapidly. Second, a statistical decision is used to solve the complexity of background updating. Experimental results show that the improved method can adapt to illumination change, and improve the precision and robustness of motion detection.