WAVELET(1)

📦
version 7.44.0 · 2 versions
, in nixpkgs unstable

💅 WAVELET(1) — the BRL-CAD girlie's guide to Haar wavelets 🎀

okay so bestie, before we even start: this is WAVELET(1), straight out of the BRL-CAD manual set. she does Haar wavelet stuff. she's iconic. let's go ✨

✨ NAME ✨

wavelet — literally performs a Haar wavelet decomposition or reconstruction. that's the whole job. she decomposes, she reconstructs, she leaves. 👑

🦋 SYNOPSIS 🦋

wavelet {-d | -r} [-1 | -2] [-t _datatype_] [-# _channels_] [-w _width_] [-n _scanlines_] [-s _squaresize_] [-W _out_size_] [-S _out_size_] [-R _avg_size_] [-D _level_] < datastream > wavelets

read it like a sentence, bestie: pick -d or -r (you have to pick one, no cap), then sprinkle on whatever flags you need, feed it a datastream on stdin, and the wavelets come out on stdout. 🫶

🌸 DESCRIPTION 🌸

so the wavelet command performs a Haar wavelet transform on a raw binary dataset that it reads from standard input, and then writes the transformed result to standard output. that's the vibe. 💅

she can go two ways:

  • decompose — break a signal down into successively coarser "average" terms plus "detail" coefficients
  • reconstruct — build the original signal back up from one of those decompositions

and the way that decomposition and reconstruction are inverse operations? obsessed. 🔁 they literally undo each other.

the transform operates on interleaved, fixed-width samples — which means it works directly on image formats like pix(5) and bw(5) by treating each pixel as one sample with however many channels that format has. no conversion era needed. 🖼️

heads up though: a two-dimensional transform requires a square dataset. square only. no rectangles allowed, sorry babe. 📐

and now the ✨dramatic warning✨ — because the Haar transform sums and differences sample values, results can overflow narrow integer types. that's not a vibe. for lossless decomposition and reconstruction, convert the data to the next larger data type first using the -t option. protect your data. protect your peace. 😭💀

💖 OPTIONS 💖

  • -d — perform a decomposition. the breaking-it-down era.
  • -r — perform a reconstruction. the putting-it-back-together era.
  • -1 — do a one-dimensional transform (per-scanline, horizontal). this is the default, so if you say nothing she's already on it. 💁‍♀️
  • -2 — do a two-dimensional transform (interleaved horizontal and vertical). the dataset must be square — do not test her on this. ⬜
  • -t _datatype_ — specify the data type of the samples, which is also the type used for the wavelet arithmetic. the letter is the first character of the C type: c (char), s (short), i (int), l (long), f (float), d (double). the default is char. 👑
  • -# _channels_ — number of values per sample point. like, use -# 3 to transform a pix(5) file so the red, green, and blue channels each get transformed individually. default is 3. 🌈
  • -w _width_ — number of samples per scanline. default is 512. 📏
  • -n _scanlines_ — number of scanlines in the dataset. default is 512. 🧵
  • -s _squaresize_ — sets the width and the number of scanlines to the same value at once, i.e. a square dataset. two-for-one, we love efficiency. ⬛
  • -W _out_size_ — width of the average image in the output. normally a full decomposition reduces this to 1, but you can specify a power-of-two value where decomposition or reconstruction should stop. the output then holds an averaged image of the given size, with the rest of the data stored as detail coefficients. 🛑
  • -S _out_size_ — same as -W, but expressed as the square size of a two-dimensional transform limit. 🟪
  • -R _avg_size_ — resume a transform from a decomposition that left off with an average image of size _avg_size_. the "pick up where we left off" flag, iykyk. ⏯️
  • -D _level_ — set the debug output level. diagnostic info about data type, channels, width, height, and limit gets written to standard error. nosy but useful. 🔍

🧚 EXAMPLES 🧚

⭐ Example 1. Display a horizontal decomposition

wavelet -d -# 3 -s 512 < img.pix | pix-fb

this decomposes each scanline of a 512x512 color pix(5) image and displays the result on a framebuffer with pix-fb(1). each scanline shows a shrinking "average" toward the left, with the detail coefficients hanging out on the right. visual slay. 🔥

⭐ Example 2. Two-dimensional decomposition with a stopping size

wavelet -d -2 -# 3 -s 1024 -S 64 < img.pix > img.wlt

performs a 2D decomposition of a 1024x1024 color image, stopping once the average has been reduced to 64x64. the output image holds a 64x64 version of the original in the lower-left corner, with detail terms filling the remainder — all ready to be inverted with:

wavelet -r -2 -s 1024 -R 64

round trip complete. she ate. 💅✨

🎀 SEE ALSO 🎀

brlcad(1), pix-fb(1), pix(5), bw(5) — go read those too, bestie, they're all part of the family 🫶

💜 AUTHOR 💜

BRL-CAD Team — the icons behind the whole thing 👑

This software is Copyright (c) 1999-2026 by the United States Government as represented by U.S. Army Research Laboratory. periodt. 📜

🍓 BUG REPORTS 🍓

found a bug? a problem? a little gremlin in the code? reports of bugs or problems should be submitted via electronic mail to devs@brlcad.org. do not suffer in silence, that's not the vibe. 📧

and that's the page, babe 💖 this was WAVELET(1) from the BRL-CAD manual set — that's the source, that's the version tag on the footer, and no, there's no date stamp on this one, don't @ me. 🌸🦋✨

📦 where this page comes from