Image to Text
Extract editable text from photos and screenshots with OCR.
Open toolLoading page…
Your file stays in this browser.
Recognize printed text on scanned PDF pages and create a searchable text result in your browser.
OCR adds an invisible text layer above scanned page images so the document can be searched and copied. Accuracy depends on scan resolution, language, rotation and print clarity.
A scanned PDF looks like text but behaves like a photograph: you cannot search it, copy from it or have it read aloud. OCR, optical character recognition, reads the printed letters and adds a text layer. This guide explains what MyFileNest's OCR PDF produces, how to get accurate recognition, and its current limits.
OCR PDF renders each page of your scan, recognises the printed text with the Tesseract OCR engine running in your browser, and builds a new PDF. Each page shows the original scan image, with the recognised text added as an invisible layer. In a PDF viewer you can then search for words to find the page they appear on, and select text to copy it.
Recognition languages currently offered are English, Spanish, French and German. A single run accepts up to 20 pages, which keeps memory use manageable on ordinary computers and phones.
| Factor | Better | Worse |
|---|---|---|
| Resolution | 300 dpi for normal print | Under 150 dpi, or phone photos taken from a distance |
| Contrast | Black text on white paper | Faded print, coloured paper, pencil |
| Alignment | Straight pages | Skewed or curled pages near a book spine |
| Layout | Single column | Tables, sidebars and text over images |
| Font | Common book and office fonts | Decorative, very small or handwritten text |
A small business has a 14-page scanned lease and needs to find every mention of "break clause" before a meeting. Running OCR PDF in English produces a searchable copy. Searching the phrase lists the three pages where it appears, and the owner can jump straight to them instead of reading every page.
To quote the clause in an email, they select the text on that page and copy it, then proofread it against the scan: OCR is very good on clean print, but a smudged character can still be misread, and legal wording should always be checked.
Use OCR PDF when you want to keep the document as a PDF and make it searchable. Use Image to Text when you want the recognised words themselves, in an editable text box you can copy or download as a .txt file, especially for accented languages. It works on images, so export the scanned pages with PDF to JPG first. If the end goal is an editable Word document, run OCR PDF first and then PDF to Word on the searchable result.
The OCR engine and language data are downloaded to your browser and recognition happens on your device. Scanned documents, which often contain signatures, addresses and ID numbers, are not uploaded to MyFileNest.
A few minutes spent on the scan saves much more time correcting text afterwards. Straighten pages with Rotate PDF if they were scanned sideways or upside down, since recognition expects text to run left to right. Remove blank pages with Delete PDF Pages so they are not processed. If you are scanning fresh, choose 300 dpi grayscale rather than colour; colour adds file size without helping recognition of black text.
For long archives, process documents in batches of up to 20 pages and name each searchable output consistently, so the searchable copies can replace the image-only scans in your filing system.
Does OCR change how the pages look? No. Each page keeps the scanned image; the recognised text is added as an invisible layer behind it.
Why does my searchable PDF look slightly different in size? Pages are rebuilt at a standard width from the rendered scan, so the file size and page dimensions can differ from the original, although the content is the same.
Can OCR read tables? It recognises the words and numbers, but not the table structure. For figures you need in a spreadsheet, check the recognised values carefully against the scan.
Make a scanned PDF searchable and copyable. Review the result before downloading, copying, or using it.
Search a scanned document. Use a representative source, compare the output with the original, and keep the version that meets the destination's requirements.
Recognition quality depends on scan resolution and contrast. Return to the retained source, adjust the relevant setting and create a fresh result rather than repeatedly processing the output.
The file is processed on this device and is not uploaded to MyFileNest. Closing or refreshing the page clears the active session.
Practical limit: Files up to 25 MB are accepted for guests. Very large dimensions, complex documents or limited device memory can affect browser processing.