Michael C. Roggemann (Pacific Defense Solutions, and Michigan Technological University), Paul Billings (Textron Systems), Jeff Houchard (Pacific Defense Solutions)
Keywords: SSA, imaging
Abstract:
The bispectrum and MFBD algorithms were developed to overcome the effects of turbulence and measurement noise in astronomical and space situational awareness systems. However, two non-ideal, but highly practical aspects of the measurements are not explicitly handled in the derivation of either of these algorithms: gray scale quantization, and the possibility of saturation. Minimizing the number of gray levels recorded allows the amount of data which must be recorded, stored, and processed to be minimized. However, gray scale quantization reduces the quality of the image reconstructions. Additionally, part of an image may occasionally saturate, causing a non-linear effect in the measurement, and also resulting in artifacts in the image reconstructions. We have investigated both of these issues, and report the results in this paper.
Here we show that the effects of gray scale quantization are smaller for the MFBD technique than for the bispectrum technique. We have also modified an implementation of the MFBD algorithm to account for saturation, and present the results.
Date of Conference: September 12-15, 2007
Track: Poster