Firefox Browser Add-ons
  • Extensions
  • Themes
    • for Firefox
    • Dictionaries & Language Packs
Log in
Add-on icon

Unified Focus version history - 4 versions

Unified Focus by Systems Research Technologies

There are no ratings yet
0 Stars out of 5
5
0
4
0
3
0
2
0
1
0
Unified Focus version history - 4 versions
  • Be careful with old versions! These versions are displayed for testing and reference purposes.You should always use the latest version of an add-on.

  • Latest version

    Version 1.3.0

    Released Nov 2, 2026 - 919.43 KB
    Works with firefox 58.0 and later
    A powerful and easily accessible all-in-one tool for photography, microscopy, astronomy, forensic analysis etc. Unified Focus is a private image-processing/forensics workbench that allows module streams (chained pipelined operations) and batch processing.

    There are 29 built-in processing modules, and you can write your own in C. Many are high-end algorithms with auto-calibration. Compiled to WebAssembly they hit bare metal and use your CPU cores in parallel: denoise, deconvolve, sharpen, stack, resize, tone-map, deskew, binarise, dither. For forensics there is noise residual, local variance, Error Level Analysis and an FFT.

    New in 1.3 — FFT module: spectrum view, notch de-screening (removes halftone dots, moiré, sensor interference and banding from a scan or photo), low pass and band stop. Error Level Analysis: finds pasted, retouched or resaved areas by re-compressing at a JPEG quality you pick; its codec is built in, so the same file gives the same map on any machine. Metadata readout: EXIF, XMP, ICC profile, GPS and thumbnail are shown when you open a file, and exports are written clean.

    Run a small server on any always-on computer and a local network becomes one shared processing pool: jobs are split across every machine on it, with per-user accounts for admins. Setup is one command (server_setup.md in the repo).

    Everything runs on your own hardware, so images never leave your computer or network. Local data is deleted after 7 days. No telemetry, no analytics, no account needed in local mode. Whole-image processing goes up to 16 megapixels.

    Release version supports single channel grey scale, colour soon to follow.
    Free and open source (AGPL-3.0)

    Source code released under GNU Affero General Public License v3.0 only

    Download Firefox and get the extension
    Download file
  • Older versions

    Version 1.2.0

    Released Oct 30, 2026 - 890.29 KB
    Works with firefox 58.0 and later
    A powerful and easily accessible all-in-one tool for photography, microscopy, astronomy, forensic analysis etc. Unified Focus is a private image-processing/forensics workbench that allows module streams (chained pipelined operations) and batch processing.

    Many high-end, optimised algorithms are included, some of which feature auto-calibration. There are 26 built-in processing modules (so far) but users can create their own custom modules. Compiled to WebAssembly they hit bare metal and use your CPU cores in parallel. Denoise, sharpen, stack, resize, tone-map and more. A forensic set is growing alongside: the Noise Residual high-pass exposes splices and unnaturally smooth synthetic regions, and the Local Variance texture map flags sharpness inconsistency across an image.

    New in 1.2 — documents and deep pixels. Open TIFF files directly: greyscale or colour, 8- or 16-bit, compressed or multi-page (a multi-page file arrives as a stack, ready to combine). 16-bit PNGs keep their full depth too — the workbench reads the extra data a browser would otherwise discard, and sets the display window for you. And two document modules: Deskew straightens a crooked scan, and Sauvola turns a page to clean black and white even when the paper is shaded, stained, or show-through from the other side. Whole folders can also be converted to PNG or TIFF in one pass.

    Networks get real accounts: admins can create individual user logins (v1.0 shared one password between machines), so work on a shared network is attributed per person. Whole-image processing goes up to 16 megapixels.

    Optionally, run a small server on any always-on computer (setup is one command, see the repo's server_setup.md) and turn a school lab, library, hospital department or club into one shared processing network: jobs submitted on any machine are split into pieces and crunched across all of them at once.

    Everything happens on your hardware so your images never leave your computer/network, and local data is cleaned up automatically after 7 days. No telemetry and no analytics — local mode needs no account at all.

    Release version supports single channel grey scale, colour soon to follow.
    Free and open source (AGPL-3.0)

    Source code released under GNU Affero General Public License v3.0 only

    Download file
  • Version 1.1.0

    Released Oct 18, 2026 - 840.4 KB
    Works with firefox 58.0 and later
    New in 1.1 — DICOM. Open medical images (.dcm) directly: window/level presets and sliders bring 8-16 bit data onto screen, an optional de-identify step strips patient details before anything is stored, and processed images can be written back as anonymous DICOM or exported losslessly as TIFF.

    Networks get real accounts: admins can create individual user logins (v1.0 shared one password between machines), so work on a shared network is attributed per person. Whole-image processing now goes up to 16 megapixels.

    Source code released under GNU Affero General Public License v3.0 only

    Download file
  • Version 1.0.0

    Released Oct 13, 2026 - 518.68 KB
    Works with firefox 58.0 and later

    Source code released under GNU Affero General Public License v3.0 only

    Download file
Go to Mozilla's homepage

Add-ons

  • About
  • Firefox Add-ons Blog
  • Extension Workshop
  • Developer Hub
  • Developer Policies
  • Community Blog
  • Forum
  • Report a bug
  • Review Guide

Download

  • Download Firefox
  • Windows
  • macOS
  • iOS
  • Android
  • Linux
  • All

Latest Builds

  • Nightly
  • Beta

Firefox for Business

  • Enterprise

Community

  • Connect
  • Contribute
  • Developer

Follow

  • Instagram
  • YouTube
  • TikTok
  • Bluesky
  • Podcast
  • Privacy
  • Cookies
  • Legal

Except where otherwise noted, content on this site is licensed under the Creative Commons Attribution Share-Alike License v3.0 or any later version.