Redact Images & PDFs
Hide the parts of an image or PDF you don’t want others to see. You can also mask all the phone numbers and addresses in a piece of text at once.
Load a file
Mark areas to hideDrag to draw · click to select (Shift for several) · drag to move · red × to delete
Found 0 possible face(s). Green dotted boxes are not hidden yet — click one to accept it.
Faces that weren’t found get no dotted box. Profiles, small faces, people seen from behind and faces with masks are often missed. Always check the result with your own eyes. Credits and licenses
What to mask
Text
What this tool can do
HANEMask is a free tool for hiding the parts of documents and text you don’t want others to see. No installation or sign-up is needed. Everything happens inside your browser, and your files and text are never uploaded to or stored on a server.
-
⬛Redact by dragging
Just drag over an image or PDF to hide that part. Besides rectangles and ellipses, you can draw freehand shapes. Black out, white out, pixelate or blur can be chosen separately for each area — for example, pixelate faces and black out prices.
-
🙂AI face detection
Finds faces in a photo and marks them as suggestions with green dotted boxes. Nothing is hidden automatically — you decide which ones to accept. Detection runs entirely in your browser.
-
🚫No hidden text left behind
PDF pages are flattened to images after redaction. This prevents the common mistake where black boxes are simply laid over text that can still be copied.
-
🔎Finds personal info in text
Finds and masks phone numbers, email addresses, postal codes, addresses, My Number (Japan’s 12-digit individual number), bank account numbers and more in pasted text. Except for emails, URLs and card numbers, the patterns are designed for Japanese formats.
-
🎛Choose how to replace
Replace each item with a label such as “[Phone]”, or block it out with “■■■■”. Labels are handy when readers should know what kind of information was removed.
PDFs are handled one page at a time as images, so long documents take a while. Text in the exported PDF can no longer be searched or copied. Hiding text while keeping it as text doesn’t truly hide it, so we deliberately don’t support that.
How to use it
- Load an image or PDFDrag and drop it, or use the “Choose” button. All pages of a PDF are loaded.
- Drag over the areas to hidePick a shape (Rect, Ellipse or Free) and draw around what you want to hide. For photos, “Find faces” suggests faces you can accept with a click. Click an area to select it, or Shift-click to select several. Drag a selected area to move it, press Delete to remove it, and use Ctrl+C / Ctrl+V to copy and paste (in a PDF you can paste at the same position on another page). Made a mistake? Press Ctrl+Z.
- Choose a style and saveBlack out, pixelate and so on can differ for each area. Select an existing area to change only that one. Before saving, use “Preview result” at the top right to hide the markers and check the result. PDFs are exported with all pages.
When it comes in handy
Sharing a quote or invoice with parts hidden
Black out just the amounts or client names and save as PDF.
Useful when you want to remove other companies’ names or unit prices from documents you share as references. The exported PDF consists of images, so the recipient cannot peel off the redaction to see what’s underneath.
Hiding bystanders or license plates in photos
Choose “Pixelate” or “Blur” and draw around just that part. For faces, “Find faces” can suggest them.
When posting site photos on social media or in documents, unrelated passers-by or license plates can be a problem. With many people, use “Find faces” and then “Accept all” to hide them quickly. License plates are not detected automatically, so draw around them by hand. Location and other metadata stored in the photo file can be removed with HANEClean.
Sharing meeting notes or inquiries outside your company
Paste them into “Auto-mask text” to mask contact details and addresses in one go.
If you forward customer correspondence to a partner company as is, the contact details go along with it. Paste it here and copy the masked version instead. Automatic detection is not perfect, though. Always check the output yourself.
How well does face detection work?
It finds reasonably large faces looking toward the camera. Some faces will always be missed.
Profiles, people seen from behind, downturned faces, faces covered by masks or hands, and small faces are often missed. In a test on an old group photo of 29 people (1920 px wide), it found 21 faces — which means some people were not found. Anything without a dotted box is not hidden. Even after accepting suggestions, always check the whole image yourself. It may also suggest things that are not faces; in that case, don’t accept them and press “Discard suggestions”.
Face detection also runs only inside this browser; your photo is never sent anywhere. It uses MediaPipe BlazeFace (Apache-2.0), which is licensed for commercial use. About 12 MB is downloaded the first time only.
What can’t be found automatically?
Information without a fixed format. In particular, names without an honorific cannot be found.
Items with a fixed format, like phone numbers and email addresses, are found with high accuracy. Names are found only when followed by a Japanese honorific such as 様 or さん (e.g. 山田様, 田中さん); a bare name like 山田 — or an English name such as “John Smith” — cannot be identified. The same goes for company, product and project names (company names are detected only with Japanese legal forms such as 株式会社). Phone numbers, postal codes, addresses and bank accounts are detected in Japanese formats only. Always review the masked text yourself.
The automatic detection in this tool (text masking and face detection) is purely mechanical. We do not guarantee that all personal information or faces will be detected. Always check the exported result yourself before using it. We accept no liability for any information leak or loss arising from the use of this tool. See our Terms of Use and Privacy Policy for details.