|
|
|
Application of Iterative Blind Deconvolution
|
|
|
| PROJECT GOALS: The overall goal of this project is to enhance images taken by NASA spacecraft or telescopic imaging systems.
More specifically, this project seeks (i) to test existing and experimental Iterative Blind Deconvolution (IBD) algorithms, (ii) to quantify
the limits and capabilities of different algorithms, and (iii) package the most useful algorithms in a form that is easily accessible to
NASA investigators. PROPOSED WORK: This project is a focused, two-year program consisting of four main tasks. 1. Test existing
IBD algorithms on various Hubble Space Telescope (HST) and Infrared Telescope Facility (IRTF) image data sets. 2. Implement and
test ideas that might improve IBD algorithms. 3. Incorporate appropriate IBD algorithms as library routines for an IRTF data
reduction pipeline. 4. Study the efficacy of new IBD algorithms for upcoming releases of AIDA (Adaptive Image Deconvolution
Algorithm) and build them into AIDA if appropriate. RELEVANCE OF PROPOSED WORK: Image enhancement is one arena where
the goals of the astronomer have immediate applications in many real-world scenarios. An enormous amount of work has been done in
the fields of medical imaging and military surveillance, to name two examples. In these examples, as with most astronomical imaging,
the common goals are to estimate the true image in the presence of noise and blurring (both sharpening features and detecting faint
details) while ensuring that no artifacts are produced by the image enhancement process. The products of this proposal will be
incorporated into AIDA, an open source IBD package (Hom et al. 2007). |
|
|
