Steven P. James (AFIT), Stephen C. Cain (AFIT)
Keywords: Imaging
Abstract:
Dim-object detection and characterization of geosynchronous satellites is one of the Air Force’s primary concerns in Space Situational Awareness (SSA). Space-to-Space imaging satellites are costly and require medium to long time scales for useful data collection. Telescope imaging is a more economical solution and can be employed in a much shorter period of time. Atmospheric seeing and the diffraction limit of the optical systems impede our efforts to get the resolutions needed for SSA.
The Light collected from satellites and other man-made objects tend to be highly polarized but, distant and/or small objects have low photo-counts in short exposure imagery. Recently, it has been shown that short exposure images of objects that possess spatial polarization diversity can be restored with resolutions as high as twice the diffraction-limit. Also, recent work in blind deconvolution of long exposure imagery and simultaneous estimation of the atmospheric seeing parameter have produced significantly improved image reconstruction, reducing the need for adaptive optics where they are not practical. The goal of this paper is to combine polarization diversity imaging with blind deconvolution of long exposure imagery to produce a new algorithm that gains the benefits of both methods. Simulated long exposure image data was used to test the new algorithm. The results show that resolution beyond the limit imposed by atmospheric seeing are possible when the object being imaged possesses spatial polarization diversity.
Date of Conference: September 1-4. 2009
Track: Imaging