多目标轨迹分离方法研究

为了实现多目标落点测量，建立了多目标测量方案，并且对大视场光电测量设备在像面坐标系下进行多目标分离的方法进行研究。该方法按目标运动方向、运动速度、运动加速度的变化，建立了相关性评价函数，根据评价函数判断像点与目标轨迹的关联程度，从而现实多目标轨迹分离工作；而且在分离过程中，考虑了同一目标像点在不同幅图像上的丢失情况，根据已有目标轨迹数据，使用卡尔曼滤波器补充丢失的像点，保证同一目标轨迹的连续性。仿真数据与实际数据结果表明，该方法有效地分离了多目标轨迹。满足多目标落点测量的设计要求。

Research of Recognition Method for

In order to realize the position measurements of multiple targets on the ground,a project of track recognition for multiple targets is established,and the recognition method in pixel coordinates for large area optical electronic measurement sets is researched.An evaluation function based on target motion,such as the direction,speed and the acceleration is established.According to the correlation degree between point and track decided by the evaluation function,the track recognitions for multiple targets are accomplished.By considering the losses of some points in some frames,the lost points are replenished by the track data with Kalman filter to ensure the track continuance during the course of recognition.The experimental results show that this method can recognize 50 tracks,which meets the requirements for position measurements of multiple targets.