A Non-Stellar Big Bright Object Elimination and Star Centroid Extraction Algorithm for Star Tracker

The star tracker is a highly accurate measuring instrument used in satellite attitude measurement and determination systems. Stellar images are important since they are the first key in the satellite attitude control process. In the lost-in-space case, stellar images are captured randomly by a star tracker, since the space environment does not only contain stars. Stellar images may include stars and even non-stellar objects. When the non-stellar big bright objects are present in the stellar images, the attitude calculation process might be challenging. This paper proposes a nonstellar big bright object elimination and star centroid calculation algorithm for a star tracker. On a set of generated stellar images, the suggested algorithm was tested. The results demonstrate the proposed algorithm's ability to remove the non-stellar big bright object's appearance and extract the star’s position from the stellar images.