When the shipborne or airborne optoelectronic sensors sways ,drop frames or the target makes complex tactical maneuvers ,the tracking target w ill suddenly change its original trajectory betw een adjacent frames .How to effectively track the abrupt maneuvering target is a difficult problem .Firstly ,a two‐frame difference method based on SU RF feature descriptor is used for background subtraction ,and then the predicted position of the target is given by the Kalman filter .In the searching area around this position ,the tracking method of M ean shift is used to find the best matching of the target .M eanw hile ,the frame‐by‐frame prediction error covariance of the Kalman filter is obtained for determining w hether the target is maneuvering .After detec‐ting the target maneuver ,all suspicious targets in the field of view are quickly detected by the efficient sub‐w in‐dow search method based on saliency density .Finally ,the original tracking target is selected by feature matc‐hing by using the SU RF algorithm and the target position is returned to realize automatic and reliable tracking of abrupt maneuvering targets .T he experimental results show that the new system can guarantee fast and accurate tracking effect for both conventional moving targets and abrupt maneuvering targets .%当舰载或机载光电传感器晃动、掉帧或者目标做复杂战术机动时,跟踪目标在相邻帧间会突然改变原来的运动轨迹,此时如何有效跟踪突然机动目标是一个难点问题.首先利用基于组合基于(speeded up robust features ,SURF)特征描述子的二帧差分法进行背景差分,然后再利用卡尔曼滤波给出目标的预测位置,在以此为中心的搜索区域内用M ean shift跟踪方法寻找目标的最佳匹配,同时逐帧根据卡尔曼滤波的先验预测误差协方差判断目标是否出现机动.在检测到目标机动后,利用基于显著密度的高效子窗口搜索方法快速检测视场内的所有可疑目标,最后利用SURF算法进行特征匹配筛选出原始跟踪目标并返回目标位置,实现突然机动目标的自动可靠跟踪.仿真实验表明,新系统无论针对常规运动目标还是突然机动目标都能保证又快又准的跟踪效果.
展开▼
