Detecting Small Moving Objects for a Monocular Automatic Visual Surveillance System

An algorithm based on a Cauchy distribution statistical model of a scene background is presented to detect low resolution moving objects for monocular automatic visual surveillance system.A robust background subtracting based on moving object detecting approach is acquired by hypothesis test. Experimental results demonstrate the proposed algorithms can tolerate the whole or local sudden or slow changes in illumination, filter clutter noises caused by small motions in background scene, and adapt to rain or haze.The detecting results of the proposed algorithm are better than those of theGaussian statistical model or the shading model.