Centroid weighted Kalman filter for visual object tracking

In the visual object tracking, the Kalman filter presents commonly the state model and observation model uncertainty in the actual performance of Gaussian noise, so it makes the estimation of certain parameters produce errors in the model, and results in decreasing estimation precision. In order to enhance the stability of the Kalman filter, an algorithm based on centroid weighted Kalman filter (CWKF) for object tracking is proposed in this paper. The algorithm firstly uses background subtraction method to detect moving target region, and then uses the Kalman filter to predict target position, combining centroid weighted method to optimize the predictive state value, finally updates observation data according to the corrected state value. Tracking experiments show that the algorithm can detect effectively moving objects and at the same time it can quickly and accurately track moving objects with good robustness.