How to scan a document in your browser
- Photograph the document. Tap Take photo to use your camera, or drop in a photo you already have. Flat surface, good light, and stand so your shadow is off the page.
- Adjust the corners. The tool suggests where the document's edges are. Drag any corner to correct it — a magnifier appears so you can land exactly on the paper's edge.
- Pick a filter. Document mode makes the paper white and the text dark, like a real scanner. Grayscale and original are there when you want them.
- Add pages and download. Tap Add page for each sheet, reorder if needed, then download everything as a single PDF.
What it does
- Automatic edge detection (straight-line tracing plus background modeling), with a draggable four-corner crop and a magnifier to fine-tune
- Handles rotated phone photos correctly — EXIF orientation is read and applied, including on older Safari
- Document filter that makes paper white and text crisp — plus grayscale and original modes
- Multi-page: keep adding pages, reorder them, and download one PDF
- Works with your iPhone camera or any photo you already have
- Runs entirely in your browser — the photo is never sent to a server
A scanner that runs where your documents should stay: on your device
Most "free online scanner" sites work the same way: you upload the photo of your document to their server, their server processes it, and you download the result. For a lease, a passport, or a medical bill, that's a strange trade. This scanner does the same job — perspective correction, cleanup, PDF assembly — with JavaScript running in your own browser. The page you're photographing never leaves the device you photographed it with.
How it actually works
When you pick a photo, the tool looks for the document's outline two ways at once: it traces the strongest straight edges in the image (the way scanner apps do), and separately models the background from the photo's borders to find the region that isn't table. When the two agree, the crop suggestion is usually within a few pixels; when the photo is tricky — busy background, document filling the whole frame — the tool says so and starts you from a sensible default. You then correct the corners by hand — a magnifier follows your finger so you can place each corner on the paper's true edge. From those four corners the tool computes a perspective transform (the same math real scanner apps use) and re-projects the page into a clean rectangle. The Document filter compares every pixel against its local neighborhood, so paper turns white and text stays dark even when the lighting across the page wasn't even. Finally, the pages are embedded into a PDF, built entirely in the browser with an open-source library.
Getting a scan that looks scanned
The transform can fix perspective, but it can't invent detail that isn't in the photo. Three habits make the difference: light from a window rather than an overhead lamp (fewer shadows), the phone held flat above the page rather than tilted (less distortion to correct), and a contrasting surface under the paper (a white page on a white table gives the edge detector nothing to find — and gives you more corner-dragging to do).
What this tool doesn't do
It doesn't read the text. The PDF you download is a stack of images — perfect for printing, emailing, or filing, but you can't search inside it, and neither can your phone. It also doesn't organize anything: page 3 of your lease will be exactly where you put it, in your Downloads folder. If you'd rather scan once and then find things by asking — "how much was the deposit?" — that's the job of a document vault with OCR, which is what Paperlock was built for.