01
Image Compressor
Compresses images to a target quality or file size in your browser, with a live…
Image Redactor
Covers parts of an image with an opaque fill, flattens it and re-encodes to PNG, so the hidden pixels and the metadata are both gone.
Image
Nothing uploaded
No sign-up
Free, unlimited
Drop the image to redactJPG, PNG, WebP — it is read by your browser and never uploaded
Add a file to begin.
In short
Redaction fails in two well-documented ways, and both are avoidable. The first is drawing a black rectangle in a PDF or a design tool, where the box is a separate object sitting on top of intact text that can be selected, copied or simply deleted — this has leaked court filings and government reports repeatedly. The second is blurring or pixelating instead of covering, which keeps a low-resolution copy of the original and has been reversed by brute force for anything with a small search space, such as a card number or a licence plate. A raster tool avoids both: the fill is written into the pixel buffer and the image is re-encoded, so there is no layer to remove and nothing underneath. The third leak nobody looks for is the metadata — a JPEG can carry GPS coordinates, a camera serial number, a thumbnail of the unedited image and the original filename, none of which a black box touches. Re-encoding through a canvas drops all of it, and this tool lists what was in the source file and what is in the export so the claim can be checked rather than believed.
How to use the Image Redactor
- Drop in the image and press Open the redactor.
- Drag a rectangle over anything that must not be readable.
- Add as many boxes as you need; Undo removes the last one.
- Press Export redacted PNG — the fill is flattened into the pixels before the file is written.
About the Image Redactor
The Image Redactor covers parts of an image with an opaque fill, flattens it and re-encodes to PNG, so the hidden pixels and the metadata are both gone. It runs as plain JavaScript inside the page, which means there is no upload step, no queue and no size limit imposed by a server — the only limit is your own device.
What people use it for
- Blacking out an account number before sending a screenshot to support.
- Removing a face, a plate or an address from a photo you are about to post.
- Preparing a document image for publication with names covered.
Details
| Method | Opaque fill written into the pixels, then re-encoded — not a layer |
|---|---|
| Reversible | No. The covered pixels are not in the exported file |
| Metadata | PNG output from a canvas carries only IHDR, IDAT and IEND |
| Verification | The source and the export are both scanned and listed back to you |
| Privacy | The file is read with FileReader and never leaves your device |
The export samples the centre of your first box and checks it is the fill colour at full opacity before telling you the redaction held. If that check ever fails the tool says so rather than handing you a file you would have trusted.
Where this tool stops
Rectangles only, and the export is a flat image — any text layer, any transparency and any animation in the source is gone. Very large images are limited by how much memory your browser will give the tab.
Frequently asked questions
Can the redaction be undone?
Not from the exported file. The fill is drawn onto the pixel buffer and the whole canvas is then encoded as a new PNG, so the original values under each box are not present in the output at all — there is nothing to recover. Keep your own copy of the original if you still need it, because this tool cannot give it back either.
How is this different from the Pixelate Image tool?
Pixelation averages blocks of pixels, which keeps a low-resolution version of what was there. For content with a small number of possibilities — a six-digit code, a number plate — that has been enough to reconstruct the original by trying every candidate and comparing. A solid fill keeps nothing. Use pixelation when you want a face softened and still recognisably a face; use this when the information must be gone.
Does it really remove EXIF and GPS data?
Yes, and it shows you. The source file is scanned for JPEG APP segments and PNG ancillary chunks — EXIF, XMP, IPTC, ICC, embedded thumbnails — and both that list and the export’s own chunk list are printed. A canvas re-encode writes only the image header, the pixel data and the end marker; there is nowhere for the metadata to go.
Why is the output always PNG?
Because PNG is lossless, so the fill is exactly the colour it says it is and the rest of the image is unchanged. Re-encoding as JPEG would soften the edges of every box slightly, which looks wrong on a redaction and gains nothing — the file is meant to be sent, not stored by the thousand.
Can I redact a PDF with this?
Not directly. Convert the pages to images first with the PDF to Image tool, redact them here, then rebuild a PDF from the images. That route also removes the selectable text layer, which is the thing that makes PDF redaction go wrong in the first place.
Is the Image Redactor free?
Free, unlimited, and no email address required. You can use it a hundred times a day and nothing changes.
Is what I paste into the Image Redactor stored?
No. The Image Redactor runs entirely in your browser using JavaScript. Your input never reaches inputtesthub.com or any third party — open your browser's Network tab while you use it and you will see no request go out.
When would I actually need the Image Redactor?
The jobs it is built for: blacking out an account number before sending a screenshot to support. Removing a face, a plate or an address from a photo you are about to post. Preparing a document image for publication with names covered. It is built for exactly those cases rather than being a general-purpose tool with this bolted on.
When is the Image Redactor not the right tool?
Rectangles only, and the export is a flat image — any text layer, any transparency and any animation in the source is gone. Very large images are limited by how much memory your browser will give the tab. Knowing where a tool stops being trustworthy matters more than the cases where it works.
Which values should I know?
Method is Opaque fill written into the pixels, then re-encoded — not a layer. Reversible is No. The covered pixels are not in the exported file. Metadata is PNG output from a canvas carries only IHDR, IDAT and IEND. These are printed on the page itself, so you do not have to take them on trust.
How large a file can I use?
There is no limit set by us, because your file is never uploaded — the only ceiling is the memory your own browser will give the tab. Very large files simply take longer to process, and a phone will give up sooner than a desktop.
Related image tools
All 4202
Image Resizer
Resizes images to exact pixel dimensions or a percentage, with aspect ratio locking and…
03
Image Cropper
Crops images to a free selection or a fixed aspect ratio, with pixel-accurate coordinates.
04
Image Format Converter
Converts between JPG, PNG, WebP and BMP, with transparency handling and quality control.
05
JPG to PNG Converter
Converts JPG images to lossless PNG, useful when you need to edit further without adding…
06
PNG to JPG Converter
Converts PNG to JPG with adjustable quality and a chosen background colour where the PNG was…
07
PNG to WebP Converter
Converts PNG to WebP, keeping transparency and typically cutting file size by a third or more.
08
JPG to WebP Converter
Converts JPG photos to WebP for smaller files at the same visual quality.
09
WebP to PNG Converter
Converts WebP images to PNG for software and platforms that still do not accept WebP.
10
WebP to JPG Converter
Converts WebP images to JPG with a background colour for any transparent areas.