RedactOffline vs Signal's blur tool

Last updated: August 6, 2026

Signal's image editor includes an automatic face-blurring tool, plus a manual blur brush, aimed at protecting identities in photos before you send them, on both iOS and Android. RedactOffline does the opposite job: it finds and redacts text — phone numbers, addresses, card numbers — and has no face detection at all. They solve different problems and are worth using together, not as substitutes for each other.

What Signal's blur tool does well

It's built specifically for a real, common need: protecting the identity of people in a photo, especially at protests or other situations where a face is the sensitive part of the image, not the surrounding text. It runs locally on the device and works automatically — it detects faces and blurs them without you having to select anything, though you can also draw a manual blur over any other area, like a license plate or a road sign.

Where it falls short on text

It doesn't look for text at all. Signal's tool has no concept of a phone number, an address, or a credit card number — it detects faces and lets you manually blur anything else, but there's no automatic classification of sensitive text the way RedactOffline's on-device classifier provides. If the thing you need hidden in a screenshot is a name in a chat log or an account number in a spreadsheet, Signal's blur tool isn't built to find it for you.

When Signal is the better choice

• The sensitive part of the photo is a face, not text.
• You're on Android, where RedactOffline isn't available.
• You're already sending the photo through Signal and want the blur applied as part of that flow.

For text — the case RedactOffline is built for — see how to redact a screenshot on iPhone. For a photo with both a face and sensitive text in it, redact the text in RedactOffline and blur the face separately; see alternatives for more on combining tools.

Using both on the same photo

They don't overlap, so there's no conflict in using both. A practical order: redact any sensitive text in RedactOffline first, export the result, then open that exported image in Signal's editor if you also need to blur a face before sending it. Doing it in that order means the face-blur step operates on an image that's already had its sensitive text permanently removed, rather than risking a workflow where an unredacted original gets shared through a different path along the way.