Detection of Faint Objects between Streak-Like Stellar Images

摘要

In this paper we propose a novel method based on the characteristics of the object that we are interested in to distinguish it from other stars, whose images are line-shaped trails. The point-like object and streak-like stellar images become more visible after some pre-processing, and operated by boundary tracking, then they can be preliminarily identified by their form factors. The coordinates of the object on the CCD-frame can be further calculated by a two-dimensional Gaussian fitting in the original exposure. Experiment tests show that the new method can distinguish the object and other stars rapidly and precisely. The codes developed under the environment of Windows/Visual C++6.0 can be applied to both a single exposure and series of images to identify an artificial satellite or space debris in motion.