Chapter 19 · 19.4
Making a scanned PDF searchable
A scanned contract, an old letter or an invoice that arrived as a photo: you open it in Euro-Office, search for a name or amount and find nothing. You can't select or copy any text either. That's because a scan is an image of a page. To the computer, it contains no letters at all, only dots.
To turn it into searchable text, you need text recognition (also known as OCR). Below you can read what Euro-Office can and can't do, and how to get a searchable PDF anyway.
Check: is it a scan?
As an example, we'll use the scanned rental contract from Bloemisterij Het Zonnetje.
- Open the PDF in your WebOké Nextcloud.
- Click the magnifying glass on the left and search for a word you can see on the page, for example Huurprijs.
- Does Euro-Office find nothing, even though the word is clearly there? Then it's a scan.

What Euro-Office can't do
Euro-Office has no built-in text recognition. WebOké's Euro-Office deliberately works without plugins, so also without a text recognition plugin; you can read why in Plugins in Euro-Office: why WebOké doesn't offer them and what to do instead. With Edit PDF and Edit Text you can only change text that is already stored as text in the PDF, not the letters in a scan.
How to make a searchable PDF
Have the text recognition done when you scan, or afterwards on your own computer. The result is a PDF that looks exactly the same, with an invisible layer of text underneath. You then simply put it in your WebOké Nextcloud.
- Your scanner or multifunction printer: most devices offer an option such as "searchable PDF", "PDF with text" or "OCR" when scanning to PDF. Turn it on and choose the language of your document, for example English.
- Your phone: many scanning apps create a searchable PDF straight away. Look for text recognition in the app's settings.
- On your computer, if you're comfortable with technical tools: the free program OCRmyPDF adds a text layer to an existing scan, for example with ocrmypdf -l eng scan.pdf doorzoekbaar.pdf. You do need to have the English language (eng) installed.
Upload the searchable PDF to your WebOké Nextcloud (see Logging in and finding your documents) and open it in Euro-Office. Searching, selecting and copying now work.
Only need the text?
Want to use a piece of text from a scan in a letter, and is it just one paragraph? Then simply retype it in a document. For longer passages, it's quicker to make a searchable PDF using one of the methods above: you can then copy the text with Ctrl + C and paste it into your document. If you want the whole PDF as an editable document, read Converting a PDF into an editable Word file; that only works well if the PDF contains real text.
Pitfalls
- Online OCR websites: handy, but you're uploading your contract to an unknown party. With personal data (ID numbers, salaries, medical information), that's a GDPR risk. It's better to do it on your own device.
- Check the recognised text: text recognition makes mistakes, especially with small print, handwriting, stamps and poor scans. An 8 becomes a B, a 1 becomes an l. Never blindly trust amounts and account numbers.
- Scan sharp and straight: 300 dpi in black and white or greyscale gives the best recognition. A skewed photo with shadows causes lots of errors.
- Redacting sensitive data in a scan: search and redact finds nothing in an image. Mark it by hand instead, see Hiding sensitive data in a PDF.
