Objects Detecting Based on Adaptive Background Models and Multiple Cues

Computer vision is to provide computer with the ability to understand the image or video, locate and identify objects, trace their motion, interpret the relationship among them. Objects detecting is an important step in the applications of computer vision. The main process of background subtraction based objects detecting is to compare the input video frame with a specified background image or a background model. The pixels are classified into foreground where the difference between the input video frame and the background model is above a specify threshold. A typical background subtraction includes four modules those are preprocessing module, background modeling module, objects detection module and post-processing module. The experiment results show the effective and robust of the proposed moving objects detecting scheme.