利用径向和环向分布特征的星图识别方法

A star map identification algorithm using radial and cyclic features

提出一种基于径向和环向特征的全天自主星图识别方法。该方法利用具有旋转不变性的径向特征作初始匹配,而以环向特征作后续匹配,并用FOV约束进一步剔除冗余匹配。为了加快匹配搜索的速度,采用查找表的方式构建径向模式库。为使观测星图中尽可能多的星找到其对应匹配星,引入了验证识别环节。仿真实验表明,该方法在较高位置噪声水平下(噪声方差为1pixel)仍能达到97.57%的识别率,比同实验条件下的栅格算法提高3%。与传统的方法相比,该方法具有较快的识别速度(18ms)和较小的存储空间(0.344Mb)。

A full-sky autonomous star map identification algorithm based on radial and cyclic features is proposed. With radial features with cyclic invariance as initial match and cyclic features as follow-up match, the algorithm applies FOV constrain to further eliminating redundant match. In order to speed up the search process, a radial mode database is constructed through table lookup. Verification is introduced to find as many as possible matches for the measured stars in sky image. Simulation shows that this method can achieve identification rate of 97.57% while std. dev. of star position is 1 pixel, which is 3% improvement than grid algorithm under the same condition. The average runtime is short (18ms) and the memory requirement is small (0.344Mb) compared with conventional algorithms.