Talk:Discrete wavelet transform

This article is rated Start-class on Wikipedia's content assessment scale. It is of interest to the following WikiProjects:

Mathematics Mid‑priority This article is within the scope of WikiProject Mathematics, a collaborative effort to improve the coverage of mathematics on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.MathematicsWikipedia:WikiProject MathematicsTemplate:WikiProject Mathematicsmathematics Mid This article has been rated as Mid-priority on the project's priority scale. Java Low‑importance This article is within the scope of WikiProject Java, a collaborative effort to improve the coverage of Java on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.JavaWikipedia:WikiProject JavaTemplate:WikiProject JavaJava Low This article has been rated as Low-importance on the project's importance scale.

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)Reply

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)Reply

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)Reply

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 math-based 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)Reply

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/elec-eng/research/sipg/resource/warehouse.htm

82.24.171.88 (talk) 02:04, 22 March 2008 (UTC)Reply

Usually the opposite denotations are used (Mallat, Daubechies): G denotes high-pass filter and H stands for a low-pass one. Spellbound mipt (talk) 21:54, 22 June 2010 (UTC)Spellbound_miptReply

It's stated there that the basis of DFT in 4D-case is:

 1  1  1  1
 1  0 –1  0
 0  1  0 –1
 1 –1  1 –1

Wrong, it's:

${\displaystyle {\frac {1}{2}}{\begin{bmatrix}1&1&1&1\\1&-i&-1&i\\1&-1&1&-1\\1&i&-1&-i\end{bmatrix}}.}$

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)Reply

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)Reply

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)Reply

It would be helpful to have a simple diagram for IDWT, just like the existing diagrams for DWT. I came looking for hints on how to combine the different components back into the original image, but I was unable to find any. In the end, I found a nice diagram http://fourier.eng.hmc.edu/e161/lectures/figures/wavelet_filterbank.gif on this page: http://fourier.eng.hmc.edu/e161/lectures/wavelets/node7.html and learned that I need to process each of the components with a filter, then sum the results together. This knowledge can also be found on Filter_bank, a link to which already exists right next to the decomposition diagram. However, I think this is relevant enough that it should be included in this page. — Preceding unsigned comment added by Mraagh (talk • contribs) 23:23, 14 January 2020 (UTC)Reply