论文部分内容阅读
由于GEO(地球同步轨道)空间目标相对背景恒星的视运动速度较慢,使用相邻帧图像差分的方法难以自动识别并跟踪目标。基于Lucas-Kanade算法,使用在全图所有星像的邻域开窗,计算统计波门内星像的移动速度,根据目标的运动特征给定全局阈值判别的方法,实现了相邻帧短曝光图像间的GEO目标自动识别与跟踪。仿真实验表明,该算法稳健可靠,星像位移计算精度为10-3,计算时间快于0.1s,在观测数据的实时处理中有较大的应用价值。
Due to the slower apparent velocity of the GEO (geosynchronous orbit) space object relative to the background stars, it is difficult to automatically identify and track the target using the method of the adjacent frame image difference. Based on the Lucas-Kanade algorithm, using the window in the neighborhood of all the star images in the whole image, the moving speed of the star image in the statistical wave gate is calculated, and the global threshold value is given according to the motion feature of the target. GEO targets automatically identify and track images. The simulation results show that the algorithm is robust and reliable, the precision of star image displacement calculation is 10-3, and the calculation time is faster than 0.1s. It has great application value in the real-time processing of observation data.