Talk:Discrete wavelet transform
WikiProject Mathematics  (Rated Startclass, Midimportance)  


WikiProject Java  (Rated Startclass, Lowimportance)  


Contents
Inconsistent with Wavelet Article
The equations in this article are all discrete time, while the equations for discrete wavelets in the wavelet article are all continous time. —Preceding unsigned comment added by 70.225.81.131 (talk) 02:41, 13 September 2007 (UTC)
 This is right. See my comments on the talk page at wavelet on the state of the wavelet articles. Nothing major has changed about this.  This page as it is should be named „fast wavelet transform algorithm“ to account for its main content.LutzL 08:46, 13 September 2007 (UTC)
No discussion for disputed tag
Someone added {{dubious}} to the last section. That template expands to "see talk page" but it was empty when I got here. Someone with a knowledge of OCaml should confirm or deny the assertion that the OCaml code is not doing the right thing. Note that the OCaml code also appears as an example on the OCaml page, so if it (the OCaml code) turns out to be wrong, it should be changed there too. Steve Pucci  talk 14:12, 20 June 2006 (UTC)
Objective Caml code irrelevant
The OCaml code is pretty ugly and obscure, I reckon its causing more harm by being distracting than good, especially seeing as this is a mathbased article, where pseudocode would be the most appropriate form of representation. If you feel I am in error in deleting it, please give justification for reverting. Dreddlox 16:48, 28 October 2006 (UTC)
University of Bath sipg warehouse
The university of bath has a selection of wavelet descriptions which I think would be interesting. Unfortunately, the use of them is not described well (at all). What do other people think about putting in a link?
http://www.bath.ac.uk/eleceng/research/sipg/resource/warehouse.htm
82.24.171.88 (talk) 02:04, 22 March 2008 (UTC)
Strange use of denotations G and H
Usually the opposite denotations are used (Mallat, Daubechies): G denotes highpass filter and H stands for a lowpass one. Spellbound mipt (talk) 21:54, 22 June 2010 (UTC)Spellbound_mipt
Wrong matrix in "Comparison with Fourier transform" section
It's stated there that the basis of DFT in 4Dcase is:
1 1 1 1 1 0 –1 0 0 1 0 –1 1 –1 1 –1
Wrong, it's:
It's either a mistake or a bad formulation, either cases need correction. — Preceding unsigned comment added by 91.79.31.66 (talk) 17:04, 27 September 2011 (UTC)
Error arising from correcting matrix but not text
The following statement is correct for the removed matrix but not for the current. Or am I misstaken?
"the second and third waves are translations of each other, corresponding to being 90° out of phase, like cosine and sine, of which these are discrete versions"
129.247.247.240 (talk) 14:43, 7 April 2015 (UTC)
Description from the castle image could be better
According to Szeliski Book, chapter 3.5.4 wavelets produce one image that contains strong influence of vertical edges, and gradients, another in horizontal features and the last one in less frequent mixed derivatives, clearly the image has this three characteristics clearly splitted. I think that that photo description should include this information — Preceding unsigned comment added by 2A01:388:474:150:0:0:1:B8 (talk) 13:34, 9 February 2018 (UTC)