Share Email Print

Proceedings Paper

Space domain constraints for least squares restoration of images degraded by atmospheric turbulence
Author(s): Donald L. Durack; Michael T. Valley
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

The residual image is analyzed to determine a constraint which regularizes the ill-posed least squares image restoration problem. The energy in the residual image is constrained at a level which depends on the noise statistics, image degradation, and restoration method. This constraint is applied individually to different image subregions, to find different regularizations appropriate in each subregion. This defines a spatially variant restoration method, even for a spatially invariant image degradation. Least squares image restoration methods using this constraint are applied to an artificial image which has been degraded by simulated long-exposure atmospheric turbulence and random noise. Quantitative analysis of the restored image shows significant improvement with this constraint, in comparison to the usual constraint on residual energy which depends only on noise statistics.

Paper Details

Date Published: 1 September 1990
PDF: 11 pages
Proc. SPIE 1312, Propagation Engineering: Third in a Series, (1 September 1990); doi: 10.1117/12.21868
Show Author Affiliations
Donald L. Durack, New Mexico State Univ. and U.S. Army Atmospheric Sciences Lab. (United States)
Michael T. Valley, New Mexico State Univ. and U.S. Army Atmospheric Sciences Lab. (United States)

Published in SPIE Proceedings Vol. 1312:
Propagation Engineering: Third in a Series
Luc R. Bissonnette; Walter B. Miller, Editor(s)

© SPIE. Terms of Use
Back to Top
Sign in to read the full article
Create a free SPIE account to get access to
premium articles and original research
Forgot your username?