No abstract

It is currently possible for space debris to remain undetected in close proximity to satellites. Current detection methods are adept at locating a single object in space, as well as two objects that are greatly separated, but have difficulty finding a second object that is nearby. The problem is exacerbated if the second object appears much dimmer than the first object. The method proposed in this paper would work with existing methods as an additional processing step that would process areas of astronomical images around places where objects are detected. In this case the proposed algorithm is designed to look at a pixel and determine how bright an object would be if there were an object there, then applies a binary hypothesis test to determine if an object is present. In theory, this will have the greatest advantage over existing methods when the objects are very close together or the second object is very dim. The algorithm has been tested on both simulated data and on a laboratory level test and improved the probability of detection by up to 300%.

In recent work, 1 the resolution limits of two variations of chromo-tomographic hyperspectral imaging sensors were discussed. In this paper, we examine another variation of this type of hyperspectral sensor the present a reconstruction method using computed tomography to estimate a hyperspectral data cube and examine the limits of resolution. The resolution analysis is then compared with a simulation to determine the validity of the calculated resolution bounds. The simulation setup is discussed and results are presented.

No abstract