School of Instrumentation Science and Opto-electronics Engineering, Beihang University, Beijing 100191, China.
Sensors (Basel). 2019 Sep 5;19(18):3838. doi: 10.3390/s19183838.
When a missile is launched, the plume generated by the propulsion system will produce a lot of fake stars in the star image, which will affect the normal work of the missile-borne star sensor. A plume noise suppression algorithm based on star point shape and angular distance between stars is proposed in this paper, which is a preprocessing algorithm for star identification. Firstly, principal component analysis is used to extract the shape features of star points. Secondly, the authenticity of star points is evaluated based on length-width ratios. Thirdly, in two consecutive frames of star images, according to the shape features of star points, the optimal matching window is determined to achieve accurate matching of the corresponding star points. Finally, the rapid elimination of fake stars is completed by the principle of invariant angular distance between true stars. Simulation experiment results show that the proposed algorithm is quite robust and fast, and the elimination ratio is high even if the number of fake stars reaches four times more than true stars. Compared with the existing star identification algorithms, when the number of fake stars is large, the advantage of the proposed algorithm is obvious. Experimentation on actual star images verifies that the proposed algorithm can meet the requirements of spacecraft even if there are a large number of fake stars in the star image.
当导弹发射时，推进系统产生的羽流会在星像中产生大量的虚假星星，这将影响导弹载星感器的正常工作。本文提出了一种基于星点形状和星点间角度距离的羽流噪声抑制算法，该算法是一种星识别的预处理算法。首先，利用主成分分析提取星点的形状特征。其次，根据长短比评估星点的真实性。然后，在两帧连续的星像中，根据星点的形状特征确定最佳匹配窗口，实现对应星点的精确匹配。最后，根据真实星点之间不变的角度距离原则完成虚假星点的快速剔除。仿真实验结果表明，所提算法具有很强的鲁棒性和快速性，即使虚假星星的数量达到真实星星的四倍，剔除率也很高。与现有的星识别算法相比，当虚假星星的数量较大时，所提算法的优势明显。实际星像的实验验证了即使星像中有大量虚假星星，该算法也能满足航天器的要求。
