Scan a document with your phone
Photograph the sheet and we straighten it, whiten it and put it into a single PDF. The images stay in your browser — we never upload them.
Photograph the sheet, or pick an image
On a phone the button opens the camera. Put the sheet on a plain, darker surface and keep all four corners in the frame.
What exactly does this scanner do?
Four things, in this order. First it looks for the four corners of the sheet in the photo — from contrast, so it finds them when the paper is lighter than whatever it is lying on. Then it straightens the page: the trapezoid of a tilted shot is pulled back into a rectangle, which is perspective correction rather than a simple rotation. Next comes the light: we estimate which corner got more of it and make the sheet evenly white — that is what removes the shadow of your phone over the paper. Finally the finished pages go into a single PDF, in the order they stand in the strip.
Corner detection is not magic and we do not claim it is. White paper on a white table, a patterned cloth, a sheet running off the edge of the frame or a strong reflection will defeat it — in that case we offer the whole photo, say plainly that it did not work, and you drag the four points into place yourself. Two gestures, and more accurate than any guess.
Why is a plain photo not enough?
Because the tax office, the bank and the insurer ask for a PDF, not for six separate JPGs. A photo is tilted, its edge shows the table, its name is IMG_2481, and every mail client displays it differently. A PDF is one file with A4 pages, it prints, and its page order cannot get scrambled.
The three filters are three different amounts of interference. Colour keeps the stamp and the blue ink of a signature and only fixes the light. Greyscale removes the light and stretches the contrast — this is what most people mean by a scanned look. Black and white throws away every grey: an invoice or a contract becomes the smallest possible file, but a photograph or a faint pencil note is unusable in that mode.
Your ID card never leaves your phone
This tool runs entirely in your browser: decoding the photo, finding the corners, straightening, the filter and assembling the PDF. There is no upload, no server-side processing, and no log line anywhere with the name of your file in it. That is why it works in aeroplane mode once the page has loaded — and why we cannot give any of it back to you once you close the tab.
If you ask for a shareable link to the finished PDF, that is the single point at which a file reaches our server — and only if you press that button.
Frequently asked questions
Will the text in the finished PDF be searchable?
No. This tool puts an image into the PDF, not letters — text recognition (OCR) is a separate operation. If you need a searchable, copyable PDF, download the scan from here and drop it into the text recogniser: that runs in your browser too.
Why can it not find the edges of the sheet?
Because detection works from contrast. White paper on a white table, a patterned surface, a corner running out of frame or a strong glare leave nothing to find. Put the sheet on a darker, plain surface, shoot from directly above, and leave a couple of centimetres around it — or simply drag the four points onto the corners.
Can I use an iPhone HEIC photo?
Yes. HEIC photos are decoded here on your device — the first such image downloads a decoder of about 1.4 MB, once, and then it stays in your browser cache. JPG, PNG and WebP images need no download at all.
Why is the scanned sheet lying on its side?
Phones store portrait-versus-landscape orientation in a separate field (EXIF), and browsers do not apply it the same way. We always use our own rotation so the image stands the same way in the editor and in the PDF — if the other way round is still the right one, the Rotate 90° button sets it per page.
How many pages can one document have?
Twenty. That is not an arbitrary number: every page is one decoded photo held in memory, and beyond that a mid-range phone simply closes the browser tab. For anything longer, build several PDFs and join them with the PDF merger.