The Application of Kalman Filtering in the System of Image Difference Moving Objects Tracking

The difference image method is always used in detecting moving target in application visual system[1]. But there is problem that the difference image algorithm globally searches image pixels, which makes larger amount of calculation, costs a lot of time and reduces the number of frames in acquisition image in the same time. Moreover, the target will be lost easily when partly blocked. To solve this problem, this paper presents a difference image method to track based on Kalman predictor, which makes full use of the prediction function of Kalman filter to predict the possible target area in the next frame and do the image difference operation in the smaller region to find the center point of target in next frame. When doing palm tracking by the algorithm put forward in this paper in the system constructed by ARM chips S3C2410X + OV511 camera, it can avoid disturbing of shading, track stably and decrease the time-consuming greatly. Therefore, this algorithm will preferably mitigate the large amount of calculation and solve shading interference existed in traditional image difference algorithm.