基于统计特征的星图拖尾快速去除方法  

Rapid removal method of star map trailing lines based on statistical features

作　　者：王敏 常晓宇 孙康[1,2] 陈金勇 孙方德[1,2] 王港 Wang Min;Chang Xiaoyu;Sun Kang;Chen Jinyong;Sun Fangde;Wang Gang(The 54th Research Institution of CETC,Shijiazhuang 050081,China;CETC Key Laboratory of Aerospace Information Applications,Shijiazhuang 050081,China)

机构地区：[1]中国电子科技集团公司第五十四研究所,石家庄050081 [2]中电五十四所航天信息应用技术重点实验室,石家庄050081

出　　处：《国外电子测量技术》2024年第10期49-54,共6页Foreign Electronic Measurement Technology

摘　　要：现代光电望远镜技术快速发展,装备探测能力不断提高。星图中存在的拖尾现象,给天文定位处理的精度和时效性都带来了巨大的压力。星图对比度低,背景强弱分布不均,如何在去除拖尾直线噪声的同时,又保留星图的目标信息和目标定位精度,是一个具有挑战性的问题。提出一种基于统计特征的星图拖尾快速去除方法,通过星图目标信息统计规律实现对单帧图像拖尾目标直线的快速检测,利用插值技术快速完成星图拖尾直线去除,从而消除拖尾对空间目标检测的干扰,降低由拖尾产生的虚警概率,提升空间目标天文定位的精度。仿真试验表明,方法可以去除星图中绝大部分拖尾直线,同时能够保留目标信息,图像质量明显改善,比Hough变换方法具有显著优势。在单线程下,针对1024×1024大小图像,使用该方法进行单帧拖尾目标去除的时间为0.71s。可以看出,在不增加计算资源的前提下,能够大幅降低计算复杂度,提高计算效率。The rapid development of modern optoelectronic telescope technology has continuously improved the detection capability of equipment.The trailing phenomenon in star maps has put enormous pressure on the accuracy and timeliness of astronomical positioning processing.The low contrast of star maps and uneven distribution of background strength pose a challenging problem of how to remove trailing line noise while preserving the target information and positioning accuracy of star maps.This article proposes a fast method for removing trailing lines in star maps based on statistical features.By using the statistical rules of star map target information,fast detection of trailing target lines in single frame images can be achieved.Interpolation technology is used to quickly remove trailing lines in star maps,thereby eliminating the interference of trailing lines on space target detection,reducing the false alarm probability caused by trailing lines,and improving the accuracy of astronomical positioning of space targets.Simulation experiments show that the method proposed in this paper can remove the majority of trailing lines in star maps while preserving target information,significantly improving image quality and having significant advantages over the Hough transform method.Under single threading,for 1024×1024 size image,the time for removing trailing targets in a single frame using the method of the present is 0.71 seconds.It can be seen that the method of the present can significantly reduce computational complexity and improve computational efficiency without increasing computing resources.

关 键 词：星图 拖尾直线 统计特征 插值技术 目标检测 

分 类 号：TN79[电子电信—电路与系统]