Scan
Digitize documents and improve their perspective using your camera. All image processing occurs on your device.
Select Images
Or drag and drop here
Scan documents with your phone camera into a PDF
Photograph pages, straighten them automatically, clean them up and assemble a PDF — all inside your browser, with no app to install.
Phone cameras replaced flatbed scanners for most people years ago, but a photograph of a page is not a scan. It is taken at an angle, so the page is a trapezoid rather than a rectangle. It carries the desk around it, the shadow of your hand, and a colour cast from whatever light you were under. It reads as a snapshot, which is exactly the wrong impression when you are submitting a document to someone.
This tool closes the gap. It finds the edges of the page in your photo, corrects the perspective so the page becomes a proper rectangle again, and applies a filter that cleans up the lighting. Then it assembles the results into a PDF.
All of it runs in the browser, including the edge detection. There is no app to install, and — more to the point — the documents people scan are passports, contracts and medical results, none of which should be passing through a stranger's server on their way to becoming a PDF.
Nothing is uploaded
Detection, correction and assembly all happen in the browser tab. Your photographed documents never reach a server.
Automatic edge detection
The page boundary is found in your photo and the perspective corrected, so an angled shot becomes a flat rectangular page.
No app to install
Works from the browser on a phone. Nothing to download, no account, no permissions beyond the photos you choose.
How to scan a document to PDF
Photograph or select the pages of your document.
Check the detected page edges and adjust if needed.
Apply a filter to clean up the lighting.
Assemble and download the PDF.
Your documents are never uploaded
Consider what actually gets scanned: passports, bank statements, medical results, signed contracts. Every step here — finding the page edges, correcting the perspective, building the PDF — happens on your own device.
When people reach for this
Sending identity documents to an application
Visa, loan and account applications ask for a scan of a passport or licence. A corrected, cleaned page looks like a scan and gets accepted; a photograph on a kitchen table often gets rejected and asked for again.
Digitising paperwork with no scanner nearby
A signed form or a contract that came back on paper, and no flatbed within reach. The phone in your pocket handles it, provided something corrects the perspective afterwards.
Capturing receipts as you go
Photographing receipts at the moment you get them, then assembling them into one PDF for a claim, means never hunting through a wallet at the end of the month.
Copying pages from a book or a file
Reference material, an archive document, a page from a manual. Perspective correction matters most here, because a bound page never lies flat.
What actually happens to your photos
Each photo is analysed by scanic, a document-scanning library that runs in the browser. It looks for the four corners of the page in the image — the boundary between the paper and whatever it is sitting on.
Those corners are what make perspective correction possible. Because the page is photographed at an angle it appears as a trapezoid, and knowing where its four corners are allows the image to be warped back into a rectangle, as though the camera had been directly overhead. That is the single step that separates a scan from a snapshot. If the detection is wrong — a dark desk, a page that bleeds into its background — you can drag the corners yourself.
The filter you choose is then applied to the corrected image on a canvas. Grayscale removes the colour cast from indoor lighting; the black-and-white mode pushes contrast hard to make text crisp on paper that photographed grey; the enhanced mode lifts contrast and saturation more gently for pages where colour matters.
Finally the processed pages are assembled into a PDF with jsPDF, at the page size and orientation you chose. Every step — detection, warping, filtering, assembly — happens on your device.
Where it falls short, honestly
- The result has no text layer. These are photographs of pages, so the PDF cannot be searched or copied from. Run it through the OCR tool if you need the words.
- Edge detection needs contrast. A white page on a white table, or a document photographed against a busy background, will often be detected wrongly. Adjusting the corners by hand is normal in those cases, and shooting against a dark, plain surface avoids most of it.
- It cannot rescue a bad photograph. Motion blur, heavy shadow, glare from a window or a page out of focus stay that way. Retaking the photo in better light beats any amount of processing.
- Files are large. Camera images are big, and a multi-page scan inherits that. Compress the result before emailing it.
- Bound books photograph with curvature near the spine that perspective correction cannot flatten, since it corrects a plane rather than a curve.
- It is a phone-camera workflow, not a substitute for a flatbed scanner where quality genuinely matters — an archival reproduction, or artwork.
Compared with a scanner app
| This tool | Typical scanner app | |
|---|---|---|
| Where your documents go | Nowhere. They stay in the browser. | Many sync to their cloud by default. |
| Installation | None — it runs in a browser tab. | An app, with its own permissions. |
| Account required | No. | Frequently, to unlock export. |
| Edge detection | Yes, with manual adjustment. | Yes. |
| Watermarks or page caps | None. | Common on free tiers. |
| Cost | Free. | Often subscription for PDF export. |