运用自组织特征映射神经网络和模糊C均值聚类算法 ,分别对星空图像中星座的划分进行了仿真 ,得到了较满意的结果。这在一定程度上提高了天文导航计算的可靠性和准确性 ,为自动天文定位、定向计算提供了可靠的数据源信息。最后将此两种算法在聚类准确度等方面进行了比较 Using self-organizing feature mapping neural network and fuzzy C-means clustering algorithm, the division of constellation in the sky image was simulated respectively, and the satisfactory results were obtained. This improves the reliability and accuracy of the astronomical navigation calculation to a certain extent, and provides reliable data source information for automatic astronomical positioning and directional calculation. Finally, the two algorithms are compared in terms of clustering accuracy and so on