Redaction Tool

Draw rectangles over sensitive areas — license plates, addresses, names, screen content. Pixels are permanently destroyed, not just hidden behind an overlay. Everything runs in your browser.

Drag & drop a file here, or

JPG PNG WebP GIF BMP AVIF Max 50MB • Processed in browser • Nothing leaves your device

Permanent Destruction

Unlike overlays or blur effects, redaction rectangles permanently replace the underlying pixels. The original data is destroyed in the exported file — it cannot be recovered by removing a layer.

Draw Anywhere

Redact any area — not just faces. License plates, street signs, addresses, document text, screen content, phone numbers, or any sensitive information visible in the image.

Touch & Desktop

Works on both desktop (click and drag) and mobile (touch and drag). Full-resolution export at up to 4096px. Undo individual rectangles or clear all at once.

How it works

Redaction vs blur: which is safer?

Blur and pixelation can sometimes be reversed — especially light pixelation, where machine learning models have demonstrated partial recovery of faces and text. Redaction with a solid opaque rectangle is the only method that guarantees permanent data destruction. The original pixels are replaced entirely; there is no hidden layer or recoverable data underneath. For automated face detection and blurring, Scanly offers a Face Blur tool — but for maximum security, redaction is the stronger choice.

What to redact before sharing photos

Common targets for redaction include license plates, street addresses and house numbers, ID cards and passports, phone numbers on signs, screen content showing passwords or personal data, children's faces in public photos, and document text containing names or account numbers. Beyond visible content, photos also contain hidden metadata — GPS coordinates, camera serial numbers, and timestamps. Use the EXIF Checker to see what metadata your photo contains, and the EXIF Remover to strip it before sharing.

GDPR, CCPA, and photo redaction

Privacy regulations like GDPR (EU) and CCPA (California) consider images containing identifiable individuals as personal data. Publishing unredacted photos of people without consent can violate these laws. Redacting identifying features before publication is one way to reduce legal risk. For checking whether a photo reveals your location through GPS metadata, use the Privacy Score tool. To see the exact GPS coordinates embedded in a photo, try the GPS Map Viewer.

Redaction for journalists and investigators

Journalists routinely redact sources' faces, documents, and location identifiers before publication. Client-side redaction is especially important in this context — the image never leaves the device, so there is no risk of server-side interception or logging. For verifying whether a photo has been manipulated before redacting, run it through the Authenticity Checker or the ELA Scanner.

Proper image redaction requires more than just covering sensitive areas — it requires permanent pixel destruction. Overlay-based redaction (adding a black rectangle as a separate layer in Photoshop or Preview) can often be removed by opening the file in a different editor and deleting the layer. PDF redaction tools that simply draw opaque shapes over text are similarly vulnerable — the text underneath can be extracted by copying and pasting. This tool avoids these pitfalls by replacing the original pixels with solid color data at the Canvas level, producing a flat rasterized output with no recoverable layers.

For a complete privacy workflow before publishing images, start by checking what metadata your photo contains with the Metadata Editor — strip GPS coordinates, timestamps, and device identifiers. Then use this redaction tool to permanently remove any visible sensitive content like faces, license plates, or document text. If you need to verify the photo's authenticity before redacting evidence, the AI Detector checks whether the image was generated by artificial intelligence.

After redaction, verify your work by zooming into the redacted areas to confirm complete coverage — a single pixel of exposed text can be enough to reconstruct a name or number. For archiving redacted evidence images, compress them with the Image Compressor and assemble into a case file PDF using the Image to PDF converter. Explore all 92 image tools. Learn more in our articles on how to redact photos, blurring faces in photos and journalist photo safety.

Frequently Asked Questions

Is this redaction truly permanent?

Yes. When you download the redacted image, the pixels behind each rectangle are permanently replaced with the redaction color. The original data is destroyed in the output file — there is no hidden layer, no undo, and no way to recover the original content from the downloaded image.

Are my images uploaded to a server?

No. All processing happens entirely in your browser using the HTML5 Canvas API. Your image never leaves your device. No data is transmitted, stored, or logged on any server.

Can I redact things other than faces?

Yes — you can redact anything. Draw rectangles over license plates, addresses, phone numbers, document text, screen content, barcodes, signs, or any area you want to hide. Unlike face blur tools that only target detected faces, the redaction tool lets you select any region manually.

What is the output format and resolution?

The redacted image is exported as PNG at the original image resolution (up to 4096 pixels on the longest side). PNG is used to avoid additional compression artifacts that could appear with JPEG. The output file includes no EXIF metadata — it is a clean pixel-only export.

Can I undo a redaction after drawing it?

Yes, while editing. Click Undo to remove the last rectangle, or Clear to remove all rectangles. The original image is preserved in memory until you download or reset. However, once you download the redacted file, the destruction is permanent in that exported file.

How is this different from Face Blur?

Face Blur automatically detects faces using AI and applies blur or pixelation. The Redaction Tool is manual — you draw rectangles over any area you choose. Redaction uses solid opaque fills (permanent pixel destruction) rather than blur effects. Use Face Blur for quick automated face anonymization; use Redaction when you need to hide non-face content or want maximum security.