A double-threshold target detection method based on adaptive background matching

To efficiently resolve the problem of quick and correct detection of moving targets in video surveillance scene, a double-threshold target detection method based on adaptive background matching is proposed. Firstly, Mode algorithm is improved to get background image, then a reliable background image is obtained through improved buffer frame mechanism to put real-time video into buffer frame. In addition, foreground object is carried out by the two threshold values of gray difference image, low-threshold value is used for detecting the obvious target; on the basis of the rough detection high-threshold value is used for removing noise target and false target. Finally, moving target is detected accurately in video surveillance scene. The experiment on the vehicle, pedestrian and other object shows that the method can not only suppress the noise and interference of false target, but also can segment foreground target rapidly and accurately.