Share Email Print

Proceedings Paper

Iterative denoising and simultaneously reconstructing algorithm: a new approach for images degraded by multiplicative signal-dependent noise
Author(s): B. Savale; Olivier Delage; Henri H. Arsenault
Format Member Price Non-Member Price
PDF $14.40 $18.00

Paper Abstract

Images of neutron distributions provided by penumbral imaging are degraded by highly signal-dependent noise, so it is difficult to separate object information from noise and to solve the inverse problem. The Richardson-Lucy restoration algorithm is a well-known method to deconvolve images given the point-spread function of the aperture, but it is very sensitive to noise fluctuations. We present a way to denoise images by a multiresolution method, and then apply a modified version of the Richardson-Lucy algorithm twice to deconvolve and to denoise simultaneously neutron images.

Paper Details

Date Published: 1 October 1998
PDF: 10 pages
Proc. SPIE 3460, Applications of Digital Image Processing XXI, (1 October 1998); doi: 10.1117/12.323241
Show Author Affiliations
B. Savale, Commissariat a l'Energie Atomique (France)
Olivier Delage, Commissariat a l'Energie Atomique (France)
Henri H. Arsenault, COPL/Univ. Laval (Canada)

Published in SPIE Proceedings Vol. 3460:
Applications of Digital Image Processing XXI
Andrew G. Tescher, Editor(s)

© SPIE. Terms of Use
Back to Top