在基于机器视觉的空间飞行器相对定姿研究中,特征提取算法是相对定姿的基础。提出了基于数学形态学的空间飞行器特征提取算法,该算法有机地融入了空间飞行器的姿轨信息,实现了特征提取中结构元素的动态重构。仿真结果表明,该算法是解决空间飞行器特征提取的一种有效方法,尤其是结构元素的动态选取思路在相关研究领域也具有一定的参考价值。
In the research of relative pose for spacecrafts based-on machine vision, the feature extraction algorithm was the basis of relative pose. A feature extraction algorithm for spacecrafts based-on mathematical morphology was proposed, infused organically the attitude and orbit information of spacecrafts, structuring elements were reconstructed dynamically in feature extraction of this algorithm. The simulations show that this algorithm is a valid method to solve the problems of spacecrafts feature extraction. It is also especially valuable for the thought of structuring elements selecting dvnamically to use in correlative research fields.