Moving objects detection method based on Gaussian mixture model in complicated environment
Tong Liu · Microcomputer Information · 2010
In video surveillance systems, noise and shake of background can influence the detection of moving objects greatly. This paper proposed an improved moving objects detection method based on Gaussian mixture model. First, it based on the Probability of every pixel in the time-domain of each Gaussian model, which is used to update the Weight and the Background. And then in the method of background subtraction we can detect the moving object with the self -adaptive threshold. The experimental results the proposed method deals well with the influence of both noise and shake of background with strong robustness.