星图识别算法的关键是识别效率高、花费时间短。过去的几十年里，人们致力于建设搜寻数据库的方法，而且确实找到了大量的搜寻方法。为了减少星图数据库的搜寻时间，提出了一种将不同恒星星等分层分类的新技术。同时，采用这种准确快速的搜寻方法提出了全新智能快速星识别算法。基于台式机的仿真结果显示这种星识别方法和数据库搜寻方法具有较高的准确性和效率。通过数据库搜寻星特征的时间复杂度为0(n)。除此以外，因为星图像的质量决定了星图识别算法准确性的改进，因此提出一种模糊边缘检测技术来解决图像的预处理问题，这种方法对于噪声消除、星特征提取、数据库建设和匹配有重要意义。

The key of the star pattern identification algorithm is highly efficient identification with little time.In the past few decades, people committed to methods to search database, and indeed invented a large number of search methods. In order to save the time of searching the star map database, the technology of hierarchical classification of different stars is presented, meanwhile, uses this fast and accurate search method to propose new smart fast star identification algorithm. Desktop-based simulation results show that this star identification methods and database search methods with high accuracy and efficiency. The time complexity of searching star features via database is 0(n). In addition, because the quality of the star image determines accuracy of the star pattern recognition algorithm, a fuzzy edge detection technique is proposed to solve the problem of preprocessing images. This method has a great significance to noise cancellation, the star feature extraction and manufacture and match of database.