About the Image to Editable PDF
Turn a photo of a document, a scan or a screenshot into a PDF whose text is real text: you can select it, copy it, search it and edit it, instead of a flat picture of words.
Before you download, the page is shown exactly as the PDF will look, and you can click any line to fix or change its text. Everything happens in your browser; the image isn't uploaded.
How it works
The image is read with OCR (optical character recognition). For every line the tool measures where it sits, how big the letters are, their colour and whether they are bold, so the PDF puts the same text in the same place, in the same size and colour.
What isn't text is kept as it looks: coloured panels and buttons become coloured boxes, thin lines and borders become lines, and icons, logos and pictures are cut from the image and placed where they were. White text on a coloured button is read on its own, so it isn't lost.
Words the reader isn't sure of, and codes that mix letters and digits (where 1 and l or 0 and O are easily swapped), are underlined in orange on the page, so you can check them first. Urdu and Arabic are drawn with a Naskh font, joined and with every dot in place, and still copy as the right letters.
Assumptions and limits
- Works best on clear printed or on-screen text: screenshots, scans and sharp, straight photos of pages. Handwriting can't be read.
- The PDF uses Helvetica (Arial) for Latin text and Noto for Urdu, Arabic and other scripts, not the original font, so the letter shapes may differ slightly while the size, colour and weight stay the same.
- Bold is detected from the thickness of the letters; italic, underline styles and font families are not.
- One image becomes one page. Images up to 50 MB; very large ones are scaled down to about 3000 pixels on the long side to read them.
Worked examples
A screenshot of a settings page
Headings, labels and paragraphs become text in their own grey and blue; the white 'Verify' on the blue button stays white on blue; radio buttons, checkboxes and icons stay as small pictures.
A photo of a printed letter
Each line comes back as editable text at the right size, bold where the letter was bold, and the coloured note box keeps its background.
A notice in Urdu
Choose Urdu, and the lines come back in Naskh, right to left, with English lines on the same page read as English.
How to use the Image to Editable PDF
- Choose the image. A JPG, PNG or WebP photo, scan or screenshot.
- Read the text. Pick the language, then check the page: click a line to change it.
- Download the PDF. Pick A4, Letter or the image's own size and save the file.
Frequently asked questions
Can I edit the text in the PDF afterwards?
Yes. The text is real text, so a PDF editor such as Acrobat or the editors built into many apps can change it, and any viewer can select, copy and search it. It's usually quickest to fix the text on our preview before you download.
Why are some words underlined in orange?
Those are words the text reader wasn't sure of, plus codes that mix letters and digits, where a 1 and an l or a 0 and an O are easy to confuse. Click each one, compare it with the image using Show original, and correct it if needed.
Is my image uploaded to a server?
No. The image is read, laid out and turned into a PDF by your browser, on your own device. Only the text reader's language file is downloaded from our site the first time you use a language.
Does it work with Urdu and Arabic?
Yes. Choose Urdu or Arabic and the text is read together with any English on the page. Lines are drawn right to left in a Naskh font with the dots in place, and copying them from the PDF gives the correct letters.
How is this different from JPG to PDF?
JPG to PDF puts the picture itself in a PDF, so the words can't be selected or changed. This tool reads the words and writes them into the PDF as text, keeping their position, size and colour, so the page looks the same but the text is editable.
This tool runs entirely in your browser. What you enter or upload stays on your device and is never sent to our servers.