基于改进的相关滤波卫星视频抗遮挡跟踪方法OA北大核心CSTPCD

Some target tracking problems such as complex background,small size and easy to be occluded in satellite videos will affect the accuracy of tracking and even lead to tracking failure.Therefore,an improved kernel correlation filtering algorithm is proposed to solve the problem of target occlusion in satellite videos and to track the target effectively.The algorithm describes the target collectively by extracting the HOG features,LBP features and SIFT features of the target,and reduces the impact of background changes by using the fusion features.An adaptive Kalman filtering algorithm is proposed to solve the problem of target occlusion in the tracking process.The ITCI value is used to determine whether the target is occluded,and the position of the occluded target is predicted.The kernel correlation filtering algorithm is selected to meet the requirement of the real-time and accuracy of tracking.The experimental results show that the improved kernel correlation filtering algorithm solves the problem of target occlusion,performs better under the change of target background,and while improves the tracking accuracy and success rate greatly.