Image to Text OCR – Scanned PDF & Image to Text

Extract text from images and scanned PDFs: give it links to PNG, JPG, WEBP, TIFF, GIF or BMP images or PDFs and get the text back, one row per page, with line and paragraph breaks and a confidence score.

Built for teams digitising invoices, letters, forms and archives, for RAG and AI pipelines that need text from scans, and for AI agents that hit a picture of text mid-task.

What you get

• The text of every page with line and paragraph breaks kept.
• 26 OCR languages: English, German, French, Spanish, Italian, Portuguese, Dutch, Polish, Swedish, Norwegian, Danish, Finnish, Czech, Romanian, Hungarian, Turkish, Russian, Ukrainian, Greek, Chinese (simplified), Japanese, Korean, Arabic, Hindi, Vietnamese and Indonesian.
• Smart PDF handling: a PDF page that already contains text is read directly, which is exact, fast and cheaper. Only scanned pages go through OCR.
• ocr_confidence (0-100) per OCR page, so you can spot pages worth a second look.
• Bulk: many links per run, up to 2,000 pages per file, multi-page TIFFs included.

Example output row

A real row from the pre-filled example:

{
  "url": "https://raw.githubusercontent.com/tesseract-ocr/tessdoc/main/images/eurotext.png",
  "file_type": "image",
  "page": 1,
  "page_count": 1,
  "text": "The (quick) [brown] {fox} jumps!\nOver the $43,456.78 <lazy> #90 dog\n& duck/goose, as 12.5% of E-mail\nfrom aspammer@website.com is spam.\nDer ,,schnelle” braune Fuchs springt\niiber den faulen Hund. Le renard brun\n«rapide» saute par-dessus le chien\nparesseux. La volpe marrone rapida\nsalta sopra il cane pigro. El zorro\nmarron rapido salta sobre el perro\nperezoso. A raposa marrom rapida\nsalta sobre 0 céo preguicoso.",
  "method": "ocr",
  "ocr_confidence": 89.8,
  "char_count": 414,
  "language": "eng"
}

Pricing

Pay per event, no subscription:

• $1 per 1,000 OCR pages ($0.001 per image or scanned page).
• $0.40 per 1,000 PDF pages that already contain text ($0.0004), read without OCR.
• $0.001 per run start at the default 2 GB memory (Apify bills the start per GB).
• Free: empty pages, pages OCR cannot read well (confidence below 45, usually handwriting or photos), and error rows.

Examples: a 10-page scanned PDF costs $0.011. 1,000 receipts cost about $1.

You can cap the cost of a run with the maximum charge setting; the Actor stops before it.

Input

Field	What it does
urls	Direct links to images or PDFs, up to 50 MB each.
languages	The language(s) in the documents, up to 3. Default English.
maxPagesPerFile	Stop after this many pages per PDF or TIFF. Default 50.

Good to know

• Works best on printed and typed text. Handwriting is mostly not readable; such pages come back with low confidence and are free.
• The OCR engine is Tesseract 5, open source, running inside the Actor. Your files are not sent to any other service.
• Links must be reachable without a login. Links to internal or private network addresses are refused.
• A link that is not an image or a PDF, a password-protected PDF, or a file over 50 MB becomes an error row, never charged, and the rest of the list continues.
• Pick the right language: OCR in the wrong language misreads accented letters.

More tools from Rowfeed

• PDF to Text – text from PDFs that already contain text, page by page.
• PDF to Excel – the tables in a PDF as rows and an Excel file: bank statements, invoices, price lists.
• YouTube Transcript API – the full transcript of any YouTube video.
• Remove Duplicates & Merge Datasets – combine several runs or files into one clean list.
• Google Sheets Export – send the results of any run to a Google Sheet.