摘要
针对近距离空间交会对接低信噪比情况下标志灯像点的识别,提出了一种基于多帧高阶互相关和最大似然比检测的标志捕获技术。首先通过单帧自适应门限检测和两次高阶互相关处理抑制大量随机噪声,然后利用最大似然比检测自适应算法从候选轨迹中剔除大部分虚假轨迹,最后通过特殊设计的后处理操作进一步剔除残留的干扰轨迹,识别出真正的目标轨迹。实验表明,可识别的最低信噪比为0.5。
In short range space rendezvous and docking, in order to solve the recognition of the image points of beacon lamps under the circumstance of low Signal-to-Noise Ratio（SNR）, a kind of beacon capture technique based on multi-frame high order correlation and the Maximum Likelihood Ratio（MLR） detecting was presented. Firstly, through single frame adapting threshold detecting and twice multi-frame high order correlation operations, the noise points were greatly suppressed. Then most false trajectories were eliminated by MLR detecting algorithm. At last, the remaining disturbing trajectories were further eliminated and the actual target trajectories were recognized by the post-processing operations. The experimental results show that the lowest distinguishable SNR is 0.5.
出处
《光电工程》
EI
CAS
CSCD
北大核心
2006年第11期1-7,35,共8页
Opto-Electronic Engineering
关键词
目标识别
目标检测
交会对接
信噪比
Target recognition
Target detection
Rendezvous and docking
Signal-noise ratio
