Research on Moving Object Detecting and Shadow Removal

The paper uses new methods to detect moving objects under different illumination environment. Improved on-line Gaussian Mixture Model combines background subtraction method is used to update the background image under gradual illumination changes environment and color gradient combines morphologic method to update the background image under sudden illumination changes environment. After analyzing the brightness and sensitivity characteristics of shadow in different colorspace such as RGB,HSV,YCbCr etc. The paper points out HSV colorspace is suitable to remove visible shadow and YCbCr colorspace is suitable to remove invisible shadow. The experiment shows that the system is effective and feasible.