How to Redact Photos — Permanently Hide Sensitive Content

You screenshot a conversation to share with a colleague, then realize your email address is visible in the corner. You photograph a parking violation but your own license plate is in the frame. You need to submit a document scan, but half the page contains information the recipient shouldn't see.

The instinct is to blur it. Maybe drop the image into a photo editor and smudge it until the text looks unreadable. The problem: blur is cosmetic, not destructive. Research has shown that lightly pixelated text can be reconstructed using neural networks trained to reverse the distortion. Gaussian blur at moderate strength leaks edge patterns that machine learning can exploit. If the content matters enough for someone to try, blur might not protect it.

Redact Photos Free

What Redaction Actually Does

Redaction replaces the original pixels with a solid opaque fill — typically black, though any color works. The distinction from blur is fundamental: there is no transformation to reverse. The original pixel data does not exist in the output file. It is gone the same way erasing a whiteboard removes the marker — the information is not encoded, hidden, or recoverable. This is why government agencies, legal departments, and journalists use redaction rather than blur when handling classified or sensitive material.

Scanly's Image Redaction Tool works entirely in your browser. You load an image, draw rectangles over the areas you want to hide, and download the result. No server upload, no account, no watermark. The processing uses the HTML5 Canvas API, which means your photos stay on your device from start to finish.

When You Need Redaction

The obvious cases — hiding faces, license plates, addresses — get the most attention. But the list of things that leak identity or sensitive data in photos is longer than most people realize.

Screen content. Screenshots of emails, chat messages, dashboards, or code editors frequently contain credentials, internal URLs, API keys, or colleague names in notification bars. A single visible Slack message in the background can expose an internal conversation.

Documents and IDs. Scanning a contract, receipt, or government form often means including information from adjacent fields. A medical bill might show a diagnosis code. A lease agreement might show a Social Security number on the same page as the section you need to share.

Reflections and backgrounds. Reflective surfaces — monitors, glasses, windows — can show content you didn't intend to share. Whiteboards in the background of office photos frequently contain project plans or credentials. Street photos might include house numbers, business names, or parked cars with visible plates.

Metadata beyond pixels. Even after redacting the visible content, your image file might still contain EXIF metadata — GPS coordinates, timestamps, camera serial numbers. Scanly's redaction tool exports without metadata, but if you're using another tool, run the result through an EXIF Remover before sharing.

Step by Step — Redacting an Image

Open the Redaction Tool and upload your image (drag-and-drop, paste from clipboard, or click to browse). The image appears on a canvas where you can draw redaction rectangles.

Click and drag to draw a rectangle over the area you want to hide. The rectangle fills with a solid color — black by default, adjustable via the color picker. You can draw as many rectangles as needed. Each one permanently replaces the underlying pixels in the exported file.

Made a mistake? Click Undo to remove the last rectangle, or Clear to start over. The original image stays in memory until you download or reset, so you can experiment freely.

When you're satisfied, click Download. The tool exports a PNG at full resolution with the redacted areas permanently destroyed. The output file contains no EXIF metadata — it's a clean image with nothing hidden in the file structure.

Redaction vs Blur vs Face Detection

Three tools, three use cases. Here's when each one makes sense.

Redaction is the strongest option for any content where reversal would be damaging — legal documents, medical records, credentials, or anything where "good enough" privacy isn't good enough. It's manual, which means you decide exactly what gets hidden.

Face Blur is faster when you specifically need to anonymize faces. The AI detects faces automatically and applies blur or pixelation. It saves time when you have group photos with dozens of faces, but it only targets detected faces — it won't catch a name on a badge or a phone number on a whiteboard.

Pixelation (available in the Image Filters tool) applies a mosaic effect to the entire image or selected regions. It's visually distinctive — everyone recognizes a pixelated area — but it's the weakest form of concealment. Low-resolution pixelation on text has been reversed in published research.

For maximum security, combine redaction with metadata removal. Redact the visible content, then verify the file is clean with the Privacy Score checker.

Common Mistakes

Redacting on the wrong layer. Some photo editors let you draw shapes on a separate layer above the image. This looks like redaction but the original image layer is intact — anyone who opens the file in the same editor can delete the redaction layer and see everything underneath. Scanly's tool avoids this entirely: the redaction is baked into the pixel data of the exported image.

Missing content at the edges. Draw your rectangles slightly larger than the content you're hiding. Text rendered with anti-aliasing bleeds a pixel or two beyond its visible boundaries, and a tight crop might leave enough of a letter edge for context clues.

Forgetting about thumbnails. Some image formats (particularly JPEG) embed a thumbnail — a smaller copy of the image stored in the EXIF data. If you redact the main image but the file retains the original thumbnail, the unredacted version is still accessible. Scanly's export strips all metadata, but if you're using other tools, check with a Thumbnail Mismatch Scanner.

Partial redaction of serial numbers. Covering half a license plate or half a document number is often not enough. Databases can be searched with partial matches, and the visible portion might narrow the possibilities enough to identify the full value. Redact the entire identifier or don't bother.

Redaction for Specific Scenarios

Real estate listings. Agents photograph properties with personal items visible — family photos on walls, mail on counters, children's artwork with names. Redacting these before listing protects the seller's privacy without retaking the shot.

Bug reports and support tickets. Developers attach screenshots showing errors, but the UI might display customer data, authentication tokens, or internal system paths. Redact before attaching to a ticket that might be visible to contractors or third-party support tools.

Journalism and investigations. Sources who provide photographic evidence need protection. Redacting identifying details — badges, uniforms, tattoos, distinctive jewelry, background landmarks — can protect a source's identity when the story goes public. Pair this with EXIF removal to strip GPS coordinates and device fingerprints.

Social media. Sharing photos of your neighborhood, workplace, or daily routine builds a detailed picture over time. Redacting house numbers, street signs, and car plates in even a few posts makes it significantly harder for someone to pinpoint your location from your feed.

One Click Away from Permanent

Redaction is irreversible by design. That's not a limitation — it's the entire value proposition. Every other concealment method leaves a trail: blur preserves edge frequencies, pixelation preserves color averages, overlays preserve the original layer. Redaction destroys the data and replaces it with nothing.

The tool is free, runs in your browser, and doesn't touch a server. If the content is sensitive enough that you're thinking about hiding it, it's sensitive enough to do it properly.