Object tracking with a background subtraction and inter-frame subtraction
Hou Wen-xi · Microcomputer Information · 2009
In this paper, a background subtraction combining with inter-frame subtraction is proposed. The moving objects can be effectively extracted with the mixed algorithm. And a Kalman predictor is used to estimate the object ’s positions in the image series . The area of object can be got by the result of Kalman predictor. Then an object model is used to find the best location of the object in the area. And a dynamic threshold algorithm is also used to decrease the influence of the noise.