A Tracking Algorithm for Debris Cloud Fragments Produced by Different Hypervelocity Impacts through Image Processing

Abstract

The number of man-made debris is sharply increasing because of more and more human activities being carried out in space, which might seriously threaten the spacecraft performance and even lead to the system instability. In order to guarantee the safe operation of spacecraft, it is necessary to make an in-depth investigation on debris clouds produced by different hypervelocity impact (HVI) experiments and make an accurate damage estimation based on debris characteristics. The interest of this paper is to discuss the main properties of debris clouds that play a central role in damage estimation of the rear wall of Whipple shield protective structure. Those properties include the impact velocity, the evolution velocity, the diameter of the projectile and the incident angle of debris to the rear wall. The purpose of the paper is to track debris fragments and construct a simple trajectory model by employing 2D image processing techniques, and make a damage estimation of debris clouds through different experimental results comparison. More specifically, by using the mean shift algorithm, image noise are greatly eliminated while reserving edges. Then, in terms of <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$K$</tex-math></inline-formula> -means algorithm, the overlapping which occurs in some frames is mostly handled and fragments of the underlying images are successfully segmented. Moreover, by utilizing the scale invariant feature transform (SIFT) technique, the image matching is fulfilled, and the trajectory is further established. Finally, based on the established trajectory, the comparison of debris clouds generated by different HVI experiments is provided, through which an accurate damage estimation is formulated. Experimental results demonstrate the effectiveness and validity of the proposed algorithm via image processing.