Talk:Deconvolution
Some online research (eg his obituary) indicates that A. Lindo Patterson was another colleague of Wiener's involved in the application of the concepts of (de)convolution. He was working in X-ray crystallography, probably at the same time or earlier than Robinson's work in seismology. Any information on this topic would be appreciated. This article needs broader information on the history of deconvolution, especially outside of seismology (my own bias). Gwimpey 21:50, 10 April 2006 (UTC)
Deconvolution and NMR
I've seen several references to deconvolution in reference to Nuclear magnetic resonance. Someone with some knowledge on the topic might want to add a little info. --Tsuji 00:23, 17 August 2006 (UTC)
New software link
This software is mainly developed for astronomical image processing:
- SImg (GPL'd software) —Preceding unsigned comment added by 217.114.211.20 (talk) 17:46, 27 January 2008 (UTC)
Nice example using Richardson-Lucy adaptive restoration
Clarkvision_com Saturn Photo Gallery 1
His homepage says:
All images, text and data on this site are copyrighted. They may not be used except by written permission from Roger N. Clark. All rights reserved.
- perhaps I shouldn't have copied that - oops!
Do you think it would be worth asking for permission to use his images ? Or at least link there ? He seems quite an authoritative source IMO - FWIW. --195.137.93.171 (talk) 02:16, 7 March 2008 (UTC)
{{reqdiagram}}
What kind of diagram is required? Please re-add this template with more details about what is wanted. thanks --pfctdayelise (talk) 12:18, 27 July 2008 (UTC)
Is "s(t) = e(t) * w(t)" correct?
I'm not an expert in this field, but I think "s(t) = e(t) * w(t)" is incorrect, as it is the functions that are convoluted, not the function values. IMHO the equation should be replaced by "s(t) = e*w(t)" or possibly "s(t) = (e*w)(t)". Dendropithecus (talk) 18:54, 9 February 2009 (UTC)
statistics of distributions
In general if you have a random variable Z that is a sum of two random variables X and Y, the probability density function of the distribution of Z will be the convolution of those of the distributions of X and Y. I have seen references to using deconvolution to estimate the distribution of the components from the distribution of their sum (and presumably either the exact or estimated distribution of the other components, or some assumptions about their nature.) Unfortunately I don't know enough about the subject to write any useful encyclopedic information about it. 200.125.112.113 (talk) 18:35, 16 July 2009 (UTC)