Scan to PDF

How to use Scan to PDF

Key features of Scan to PDF

About Scan to PDF

Scan to PDF turns your phone or laptop camera into a simple document scanner. The live video runs in a viewfinder on the page; each capture copies the current frame to a canvas, stores it as a JPEG, and shows it as a thumbnail. When you convert, pdf-lib builds a new document with one A4 page per image and hands you the PDF.

It is useful for sending a signed form back by email, keeping a copy of a receipt, or turning a few photographed pages into a single file. Two limits: there is no edge detection, cropping, rotation or contrast filter, so frame the page well before you capture; and images added from the file picker are re-encoded as JPEG unless they are already JPEG or PNG, so lossless formats lose their compression.

When to use Scan to PDF

A few sheets of paper, a laptop and no scanner: open the camera, hold each sheet up to the webcam or lay it flat under a phone, and capture them one after another. Each capture becomes its own A4 page and the whole set downloads as one PDF.

Photos taken earlier with the camera app are usually sharper than a frame from the live video, because the stream is requested at 1920 x 1080 while a camera app saves a much larger picture. Add them with Choose Images / Scans; they can be mixed with live captures in one document.

A landscape capture, such as a wide receipt or a double-page spread, is scaled to fit the width of the portrait A4 page and centred, leaving white space above and below. There is no landscape page option, so capture single pages upright when you can.

For images that should become pages at their own size, with no margin and no A4 frame, use JPG to PDF or PNG to PDF instead. To make the scanned text searchable, run the finished file through OCR PDF. A page that came out sideways or in the wrong place can be fixed afterwards with Rotate PDF or Reorder Pages.

Tips and troubleshooting

• The alert "Camera access unavailable or permission denied" appears when you refuse the permission prompt, when the device has no camera, or when the page is not served over HTTPS, since browsers only allow camera access on secure pages. Choose Images / Scans still works in all three cases.
• If Capture Page does nothing, the video has not started yet; wait until the viewfinder shows a picture. Each capture is the current video frame saved as a JPEG at quality 90, at the resolution the camera delivered.
• The image picker has no size limit and accepts any image type. Pages are held in memory as data URLs until you convert, so many large photos use a lot of memory.
• When you add several files at once they are read in parallel and can land in the list in a different order from the picker, so check the P1, P2 labels on the thumbnails before converting.
• A phone photo whose orientation is stored only in its EXIF tag can come out sideways, because pdf-lib places the stored pixels and ignores the tag. Turn the page afterwards with Rotate PDF.
• The file name is scanned_document_ plus the date in UTC, so late in the evening or early in the morning it can differ by a day from your local date. Rename it after download if that matters.
• A file the browser cannot open as an image, or a photo that was cut off or damaged, is not added: a line under the buttons says "Could not read" and names it. The pages already in the list stay.
• Each photo is fitted onto an A4 page with a margin around it. If the result has more white border than you want, Crop PDF trims a preset amount from every side of every page.
• The PDF holds pictures of the pages, so its words cannot be searched or copied yet. How to make a scanned PDF searchable shows how to add a text layer with OCR PDF.

How Scan to PDF works in your browser

Open Camera Scanner calls getUserMedia, asking for the rear-facing camera at 1920 x 1080 when available, and attaches the stream to a video element. Capture Page draws the current frame onto a canvas of the video's size and exports it with toDataURL as a JPEG; Choose Images / Scans reads each file with FileReader. Both produce data URLs that go into one array and are shown as thumbnails.

Convert Scans to PDF stops the camera and creates an empty document with pdf-lib, the only library this page loads. For each data URL it reads the MIME type from the prefix: image/png is embedded with embedPng, image/jpeg with embedJpg, and anything else is decoded by the browser, drawn onto a white canvas, exported as JPEG at quality 90 and embedded. Each image goes onto a 595.28 by 841.89 point page. pdf-lib saves the document to bytes, the bytes become a Blob, and Download File saves it through an object URL. The stream, the captures and the PDF exist only in the tab; no request carries any of them.

Frequently asked questions