Star extraction method for star sensor at a high angular rate

When a star sensor slews at a high angular rate, the stars in the sensed image are smeared out over many pixels, which results in a decreasing signal-to-noise ratio (SNR). Therefore, the faint stars are not capable of being detected, and the centroidal precision of the bright stars decrease when using traditional image threshold segmentation algorithm and centroid method. In this paper, we propose a method to obtain accurate star location for star sensor at a high angular rate. We first utilize Laplacian to segment the star image, and then each pixel is marked after it is compared with the neighbouring one. Finally, the centroids of stars are calculated by using target pixels, and the centroiding errors are compensated. The experiment results illustrate that the proposed technique can extract more stars than other traditional methods under the condition of high angular rate, and also can improve the locating precision of star effectively.