一种改进的快速星图识别算法

A Modified Star Map Identification Based on Fast Algorithm

为解决传统算法在星图噪声加大、星等接近的情况下识别率和识别速度降低的问题,提出一种改进的星图识别算法;从三方面进行改进:首先采用重叠正交网格方法对星表进行均匀化,以减小导航星库容量;其次在匹配时采用哈希查找的搜索方式,以提高匹配速度;最后在识别匹配中加入星等判断环节,当星等噪声大、相对星等小时不采用星等信息,只采用角距进行匹配,同时引入了校验环节以降低冗余和误匹配;仿真实验表明,改进的算法对位置和星等误差比传统的星图识别算法具有更好的鲁棒性,且实时性得到了很大的提高。

A modified study of all--sky star map recognition algorithm has been carried out. Firstly, a method of overlap orthogonal grids was presented to select guide stars. Then star angular distances are arranged according to their value increasing, and the star angular distances list is divided into several partitions. In order to speed up the star distances indexing, both Hash and Binary Search are used to match the star angular distance. Traditional methods were invalid when the star magnitude of the observed star are close to each other or the magnitude noise is serious. So an modified algorithm was put forward to solve the problems. Finally, simulation shows that the algorithm proposed in this paper is superior than the traditional triangle algorithm. Also, this algorithm has the advantage in terms of robustness against the star position noise and star magnitude noise.