What this tool does
This is the highest-volume search in the whole PDF category, and the reason is unglamorous: people photograph a document, or scan a stack of pages, and need one file rather than forty attachments. Doing it well is mostly a geometry problem, because a phone photo is not a page and the difference between a legible scan and a tiny thumbnail is entirely in the numbers.
How it works
A PDF page is measured in points, one point being one seventy-second of an inch, and resolution is measured in dots per inch. Those two facts decide whether your scan is sharp. A4 is 8.268 inches wide, so a scan filling the page width needs 1240 pixels at 150 dpi and 2480 pixels at 300 dpi. A modern phone produces 4032 pixels, which is comfortably enough for 300 dpi and 487 dpi if you use only the middle of the frame. The tool computes this for the size and resolution you pick, so you can see whether a photo is sharp enough before you build the PDF rather than after you print it.
The alternative to scaling down is cropping, and it is a genuine choice. Fitting the whole photo onto a page forces a 4032-pixel image into 8.268 inches of paper, which works but throws away resolution and letterboxes the result. Cropping to the document first keeps more detail on the page and is what you want for a receipt or a form. Both modes are offered, and the difference is shown as a number rather than left to be discovered on paper.
Phone photos need one piece of handling that scanned images do not. Cameras record which way up the image is in a metadata tag rather than by rotating the pixels, so a portrait photo taken sideways is stored as a landscape image with a flag attached. A converter that ignores the flag puts the page on its side, and the usual response is to rotate it afterwards by hand, once per file. This tool reads the orientation and applies it to the pixels before the page is created, which is the same handling the image tools on this site use for the same reason.
There is a limit worth stating plainly. A PDF built from photographs has no text layer, so it cannot be searched, its text cannot be selected and a screen reader sees nothing. That is a property of a photograph of a document, not a fault in the conversion, and the only thing that changes it is optical character recognition, which needs a rendering engine and a language model this site does not load. Where the file has to be searchable, it has to start as a text-based PDF rather than as images.
Worked example
You have eight photos of a printed invoice taken on a phone and need one PDF small enough to email, at a resolution that will still print.
- Add the eight images in order; each contributes one page
- Choose A4 and 150 dpi, which is 1240 pixels across the page width
- The tool reports the effective resolution of each photo against the page it will fill
- If a photo is sharp enough, fit it; where only part of the frame is the document, crop instead to keep the detail
An eight-page A4 PDF in the right order, sized to the page rather than shrunk to fit a thumbnail, with a known resolution and no upload.
Accuracy and limitations
- The result has no text layer. It cannot be searched, and the text cannot be selected, because the page is a photograph rather than text.
- Resolution is decided by the source image. Scaling a small image up to fill A4 does not add detail, it only makes the same pixels larger and larger.
- Printing in colour from a CMYK JPEG can shift the colours, because those files carry a different colour model than the page assumes.
- Very large phone photos mean very large PDFs. If the file is going to be emailed, check the size and compress the images first if needed.
Frequently asked questions
- How many photos can I put into one PDF?
- There is no hard page limit, so you can add as many images as you have. The practical limit is memory: every image is held in the tab at once, so a few dozen large phone photos is comfortable while a hundred full-resolution ones may exhaust the browser's memory.
- What resolution do I need for a printable PDF?
- 150 dots per inch is fine for reading and sharing, and 300 dpi is the standard for anything that will be printed. On an A4 page, 150 dpi needs 1240 pixels across the width and 300 dpi needs 2480, so a modern phone photo is comfortably enough for both.
- Why is the text in my converted PDF not selectable?
- Because the page is a photograph of a document, so the only text in the file is part of an image. Nothing can be selected or searched until optical character recognition adds a text layer, which is a different operation and needs a heavy engine. Where a searchable file is required, it has to start as text-based rather than as photographs.
- Should I crop my photos before converting them?
- If the document does not fill the frame, yes. Cropping to the document first puts more of the photo's resolution onto the page, so the result is sharper. If the document already fills the frame, fit the whole image and let it scale to the page instead.
- Why did one of my photos end up sideways?
- That happens when the orientation information in the image is not being read. This tool reads it and turns the pixels before building the page, so portrait photos stay upright. If you see a sideways page, it means the file had no usable orientation information at all and needs rotating by hand.
- Are my images uploaded to be converted?
- No. The images are decoded, sized and placed into the PDF entirely in your browser. That matters here more than for most tools, since phone photos of documents are often receipts, statements and other things you would rather not hand to a stranger's server.