论文部分内容阅读
为了解决星载光电跟踪系统对连续帧图像快速识别、准确定位的要求.在深入分析连续图像序列特性和星载光电跟踪系统特点的基础上,利用光电跟踪系统转台的角度变化信息,对图像序列进行帧间差值运算以获得目标残差图,并基于残差图和原始图像信息提出了一种新的目标识别定位算法,大大地减小了运算量.结合中值滤波和自适应波门跟踪算法,实现了运动目标的快速跟踪定位.通过对目标实测图像序列的实验,结果证明,该算法具有快速、稳定等优点,能满足星载光电跟踪系统实时图像跟踪的要求.
In order to solve the requirement of fast recognition and accurate positioning of continuous frame images by spaceborne electro-optical tracking system, based on the analysis of the characteristics of continuous image sequences and the characteristics of spaceborne electro-optical tracking system, using the information of angle change of photoelectric tracking system turntable, A new target recognition and localization algorithm is proposed based on the residual image and the original image information, which greatly reduces the computational complexity.With the combination of median filtering and adaptive wave-gate Tracking algorithm to achieve the fast tracking of moving targets.Experimental results of the target measured image sequences show that the proposed algorithm has the advantages of fast and stable and can meet the real-time image tracking requirements of the satellite-based photoelectric tracking system.