Gray Imaging Extended Target Tracking Histogram Matching Correction Method

Abstract Due to the extended target gray color information is not rich, a series of features that target and background are overlaps make tracking point happen drift, we put forward correction extended target tracking method based on the mean shift histogram. Based on the pixel position and pixel value histogram mode as a target mode; Similarity adopts the distance defined by Bhattacharyya coefficient; By mean offset iterative matching area search; Adopt histogram matching factor amended. By establishing kalman prediction tracking strategy solving the target tracking problem when occluded. Experiments results show that, adopting the algorithm, tracking drift from originally dozens of pixels volatility reduced to accidentally have 1 pixel dithering.