Convert images to a PDF

Photographing a document on a phone

On a phone there is a “Take a photo instead” button under the drop zone. It opens the camera, and each shot joins the ones before it, so a five-page document is five photographs and then one PDF.

It is worth saying what that button is not. It does not ask for camera permission, because it does not use the camera API: it hands off to the camera app your phone already has, which returns a photograph the same way choosing one from the gallery does. This page never sees a video stream, and the site’s permissions policy still denies camera access on every route.

The button only appears where a camera is plausible. On a laptop it would open a webcam pointed at the ceiling, so it is not offered there.

Combine receipts, scans, screenshots, or photographs of a signed form into one PDF without uploading them. Each image becomes a page, in the order shown in the file list, and the list can be rearranged by dragging a card or by moving it with the keyboard before you build the document.

A JPEG or PNG is not recompressed on the way in. A JPEG is embedded with its original bytes intact, so a 125 KB photograph adds roughly 125 KB to the PDF and looks exactly as it did before. The practical consequence is that thirty phone photos make a PDF around the size of thirty phone photos. When the file has to clear an email limit, resize or compress the images first, then assemble them here.

Each page takes the pixel dimensions of its image, and PDF measures pages in points at seventy-two to the inch. A 960 by 1440 screenshot therefore produces a page about 13 by 20 inches, and a full-resolution camera photo produces something far larger still. Readers and printers scale a page to fit the paper, so it prints, but a mixed set of images yields a document of mixed page sizes rather than a tidy stack of A4. Resizing the images to matching dimensions beforehand is what makes the pages consistent.

The declared file type chooses the decoder, so an image saved as PNG but renamed with a .jpg extension is rejected with an error instead of being embedded as something it is not. HEIC from an iPhone, WebP from the web and GIF are decoded on your device, since a PDF holds JPEG and PNG pictures only. A HEIC or WebP photograph with no transparent pixel is saved as a JPEG at quality 0.92, a second lossy pass that keeps the page small. A GIF, a picture with transparency and a flat graphic such as a screenshot stay lossless PNG, which can be many times the size of the original file. To choose the quality yourself, convert those pictures to JPG in the image tools first.

The output has no text layer. A page built from a photograph of a document is a picture of that document, not searchable text, so run the result through OCR if the words need to be findable. Every image stays on your device throughout, which matters for the passports, payslips, and medical letters that people most often need to turn into a single PDF.