You click into a PDF, try to highlight a word, and nothing happens. The cursor just slides over the page like it's a photograph. That's because it is a photograph. Your PDF has no text layer, just a bitmap image of the page, and until you run OCR image PDF text recognition on it, you can't select, search, or copy a single character. Here's how to fix it, from the fastest free method to the proper batch-processing setup.

TL;DR

To OCR image PDF text, use OneNote's File Printout feature for a quick free fix, an online tool like PDF24 for single documents, or dedicated PDF tools software for reliable batch processing and accurate output. The root cause is always the same: the PDF was saved as an image scan with no text layer attached.

⏱️ 13 min read ✅ 90% success rate 📅 Updated June 2026

Key Takeaways

OCR image PDF text problems always come down to a missing text layer, not a broken file.
OneNote is the fastest free fix for a single document on Windows, no extra installs needed.
Online tools work well but should never be used with confidential or sensitive documents.
Adobe Acrobat Pro gives the best accuracy, especially with the Enhance Scanned File option enabled.
OCRmyPDF is the best free option for batch processing, but requires comfort with the command line.
Setting the correct document language before running OCR makes a bigger difference than most people expect.

At a Glance

Difficulty: Easy to Medium
Time Required: 5 to 30 mins depending on method
Success Rate: 90% of users with clean scans

What Causes OCR Image PDF Text Problems?

The short answer: your scanner saved the document as a picture, not a document. When a scanner captures a page, it produces a bitmap image, basically a photo of the paper. If the scanning software doesn't run OCR during that process, or if it was turned off in the scan profile, the resulting PDF is just a container holding that image. No text. No metadata. Nothing to select.

This catches people out constantly. A PDF looks like a document. It has pages, it has what appears to be text on screen, and it opens in exactly the same viewer as a proper PDF. But there's no actual text data in the file at all. Your PDF reader is just displaying an image and has nothing to work with when you try to select or search.

A few specific things make this worse. Low scan resolution, anything below 200 DPI really, produces images that are blurry enough to trip up even good OCR engines. Skewed pages, where the paper was fed at a slight angle, cause character misreads. Heavy JPEG compression (common on older multifunction printers set to their lowest quality setting) introduces artefacts that look like noise to the recognition engine. And using the wrong language setting is a surprisingly common one: if your OCR tool is set to English and the document is in French or German, accuracy drops significantly because the engine is working against its own dictionary.

The other cause worth mentioning is the tool you're using to open the PDF. Most basic PDF readers, including browser-based viewers like the one built into Chrome or Firefox, have no OCR capability at all. They can display the image just fine, but they can't add a text layer. You need a proper PDF tool to do that. Which brings us to the fixes.

Quick check: open your PDF, press Ctrl+A to select all. If nothing highlights, or if only the images select as objects, you're dealing with an image-only PDF and you need OCR before anything else will work.

OCR Image PDF Text: Quick Fix (5 to 10 Minutes)

These two options cover most people. No paid software, no command line, just results.

Use OneNote to Extract Text from a Scanned PDF Easy

OneNote has had built-in OCR for years and most Windows users already have it installed. It won't give you a searchable PDF back, but it will give you the text content fast, which is often all you actually need.

Open OneNote
Use the desktop version, not the web app. Go to any notebook page where you want to work.
Insert the PDF as a printout
On the ribbon, click Insert, then File Printout. Select your scanned PDF. OneNote inserts each page as a separate image on the page. For a 20-page document this can take 30 seconds or so, just let it finish.
Right-click and copy the text
Right-click any page image and choose Copy Text from All the Pages of the Printout. The option might take a few seconds to appear while OneNote processes the images. Don't click away, it will show up.
Paste into Word or Notepad
Click where you want the text and press Ctrl+V. For a clean editable document, paste into Word. For plain text, Notepad works fine. Save as DOCX or TXT.
Optional: export back to PDF
If you need a searchable PDF rather than a Word file, go to File, Export, then Export to PDF in Word. The resulting PDF will have a proper text layer.

If you can now Ctrl+F and find words in the exported document, it worked. OneNote's OCR is genuinely good for clean typed text in common languages. Handwriting and low-quality scans are where it starts to struggle.

OneNote OCR does not preserve layout, columns, or tables well. If formatting matters, use Acrobat Pro instead (see below).

Use an Online OCR PDF Tool Easy

For a single document where you want a searchable PDF back (not just the text), an online tool is the quickest route. PDF24's OCR tool is free and doesn't require an account.

Go to an online OCR PDF tool
Open your browser and search for PDF24 OCR or similar. These tools let you upload a PDF and get a searchable version back.
Upload your PDF
Click the file selection area or drag and drop your PDF onto it. Wait for the upload to complete.
Set the language
If the tool offers a language option, set it to match your document. This matters more than most people realise.
Run OCR and download
Click the button to start processing. When it finishes, download the new PDF. Open it and press Ctrl+F to confirm text is now selectable and searchable.

Do not upload confidential, legal, medical, or financial documents to any third-party web service. For sensitive files, use a desktop tool.

These tools work well for typical office documents. Very large PDFs (50+ pages) may time out on free tiers.

More OCR Image PDF Text Solutions (15 to 30 Minutes)

If the quick fixes didn't cut it, or if you need better accuracy and layout preservation, this is where to go. A proper dedicated PDF tools application handles OCR image PDF text far more reliably than OneNote or a browser-based tool, especially for documents with columns, tables, or mixed content.

Use Adobe Acrobat Pro Scan and OCR Medium

Acrobat Pro DC is the industry standard for a reason. Its Scan and OCR tool does a proper job on most documents, including slightly skewed or noisy scans, and it keeps the original page layout intact. According to Adobe's own documentation, the Enhance Scanned File option can significantly improve recognition accuracy before OCR even runs.

Open the PDF in Acrobat Pro DC
If Acrobat detects unrecognised text it may prompt you automatically. If not, open the Tools pane on the right and click Scan and OCR.
Choose Recognise Text in This File
In the Scan and OCR panel, select this option. A settings dialog will appear.
Set the document language
This is the step most people skip and then wonder why the output is garbled. Set it to the actual language of the document. If it's a mix, pick the dominant one.
Optional: run Enhance Scanned File first
If your scan is low quality, skewed, or has noisy backgrounds, click Enhance Scanned File before running OCR. Choose the content type (document, whiteboard, etc.) or let Acrobat auto-detect. This pre-processing step alone can turn a bad OCR result into a good one.
Click Recognise Text and wait
Processing time depends on page count and scan quality. A 10-page document usually takes under a minute. A 100-page document might take 5 to 10 minutes.
Review with Correct Recognised Text
Back in the Scan and OCR panel, use Correct Recognised Text to step through any characters Acrobat flagged as uncertain. Fix them manually if needed.
Save the file
Go to File, Save As, and save the processed PDF. Open it fresh and test with Ctrl+F. You should now be able to search and select text throughout the document.

Acrobat Pro's OCR preserves the original page layout, so columns, headers, and tables look right. This is the main advantage over OneNote for complex documents.

Batch Processing Multiple PDFs

If you've got a stack of image-only PDFs to work through, Acrobat Pro handles this too. Open Scan and OCR from the Tools pane, then choose Recognise Text in Multiple Files. Add all your PDFs to the dialog, set the language and output options once, and let it run. Acrobat processes each file in sequence and saves the searchable versions to your chosen output folder. This is a proper time-saver if you're dealing with a backlog of scanned documents. For even larger batches or automated pipelines, see the advanced section below.

One thing to note: Acrobat Pro is a paid subscription. If you only need to OCR image PDF text occasionally, the free tools above are fine. But if you're processing documents regularly, or if accuracy and layout matter, the subscription pays for itself quickly. There are other capable PDF suites too, including Kofax Power PDF and Tungsten Automation for enterprise environments, both of which offer similar batch OCR workflows with additional compliance output formats like PDF/A.

Advanced OCR Image PDF Text Fixes (30+ Minutes)

These are for power users, IT staff, or anyone who needs to process large volumes of scanned PDFs without paying for Acrobat. OCRmyPDF is genuinely excellent and completely free. It's just not point-and-click.

Use OCRmyPDF (Command-Line, Free) Advanced

OCRmyPDF is an open-source Python tool that adds a proper text layer to scanned PDFs without altering the original images. It's well-maintained, the documentation is thorough, and it handles batch jobs cleanly with a simple script. The catch is you need to be comfortable with a terminal.

Install Python
Download Python from python.org. During installation, tick Add Python to PATH. Without that checkbox, nothing will work from the command line.
Open Command Prompt or PowerShell
Press Win+R, type cmd, press Enter.
Upgrade pip
Run: python -m pip install --upgrade pip wheel
Install OCRmyPDF
Run: python -m pip install ocrmypdf
On Windows you may also need Ghostscript installed separately. The project docs list all dependencies clearly.
Run OCR on a single PDF
Run: ocrmypdf input.pdf output.pdf
This creates a new PDF with a text layer added. The original images are untouched. The output file is what you'll use going forward.
Batch process a folder
For multiple files, a simple PowerShell loop works well:
Get-ChildItem -Filter *.pdf | ForEach-Object { ocrmypdf $_.FullName "output\\$($_.Name)" }
Run that from the folder containing your PDFs and it will process every one of them.
Set the language if needed
Add the -l flag for non-English documents: ocrmypdf -l fra input.pdf output.pdf for French, for example. Language codes follow the Tesseract standard.

OCRmyPDF is excellent for clean scans and large batches. It's what I'd reach for if I had 500 PDFs to process and didn't want to leave Acrobat running overnight.

If you see errors about missing Ghostscript or img2pdf, install those first. The OCRmyPDF docs have platform-specific instructions. Don't skip them.

For organisations that need this at scale with compliance output (PDF/A archiving, audit trails, specific language packs), commercial suites like Kofax Power PDF or Tungsten Automation are worth evaluating. They wrap this kind of workflow in a proper GUI with enterprise controls, but they're not cheap. For most small businesses and individuals, OCRmyPDF does the job without the cost.

If you're also dealing with PDFs that won't open at all, or files that appear corrupted after OCR processing, see our guide to fixing corrupted PDF files for the next steps. And if you're working with documents that need to be converted between formats as well as OCR'd, our PDF to Word conversion guide covers the cleanest ways to do that without losing formatting.

Preventing OCR Image PDF Text Problems

The best fix is not needing one. Most OCR problems come from the scanning stage, not the PDF tool, so that's where to focus.

Configure your scanner properly. This is the single most effective thing you can do. Every modern scanner and multifunction printer has a scan profile setting. Find it, and make sure the output is set to searchable PDF, not image PDF or JPEG. On most devices this is buried in the advanced settings of the scanning software or the printer's web interface. Set it once and you're sorted for every scan going forward.

Scan at 300 DPI minimum. Anything lower and OCR accuracy drops noticeably, especially for small fonts. 300 DPI is the standard for text documents. 400 to 600 DPI if you're scanning documents with very small print or fine detail. Higher than that is usually overkill and produces unnecessarily large files.

Set the correct language in your scan profile. If your organisation scans documents in multiple languages, create separate profiles for each. The language setting affects which dictionary the OCR engine uses to validate its character guesses, and getting it wrong produces garbled output that looks fine at a glance but is full of errors on closer inspection.

Archive as searchable PDF/A. If you're storing documents long-term, use PDF/A format with an embedded text layer. PDF/A is an ISO standard designed for archiving, and most proper PDF tools support it as an export option. It's a better choice than plain PDF for anything you need to find and read years from now. The Microsoft support documentation on OneNote OCR also covers some useful edge cases if you're using that as part of your document workflow.

And if you're managing this across a team, standardise the tools. One PDF application that everyone uses, with OCR-on-by-default settings, prevents the whole problem from recurring. It's a lot easier to set up one shared scanner profile than to retrospectively OCR hundreds of documents someone saved as image-only files.

OCR Image PDF Text: Summary

To OCR image PDF text, you've got four main options depending on your situation. OneNote is the fastest free method for a single document on Windows, no extra software needed. Online tools like PDF24 work well for non-sensitive files when you want a searchable PDF back. Adobe Acrobat Pro gives the best accuracy and layout preservation, especially with Enhance Scanned File enabled, and handles batch jobs cleanly. OCRmyPDF is the free power-user option for large batches via the command line.

The underlying problem is always the same: no text layer in the file. Once you've run OCR image PDF text recognition using any of these methods, test it with Ctrl+F to confirm the text is actually searchable. And going forward, configure your scanner to apply OCR at scan time so you never have to fix this retrospectively again. If you're also dealing with related document workflow issues, our comparison of PDF tools for Windows covers which applications handle OCR, editing, and conversion best in 2026.