By adopting appropriate threshold

the algorithm of threshold centroid can get better results. The optimization threshold varies with influences of Signal-to-Noise Ratio

star point location and optical structure. Threshold obtained by traditional method can not satisfy the request of dynamic optimization for the centroid algorithm. By analyzing many kinds of influencing factors

a method of obtaining the optimization threshold based on luminosity of star spot was advanced. By studying the relationship between optimization threshold and luminosity of the spot

a curve model was produced. Based on luminosity of the spot

the optimization threshold was obtained by interpolation. The simulation results show that the algorithm can obtain the optimization threshold. Compared to traditional threshold

it can raise centroid accuracy with different Signal-to-Noise Ratio.