Image to Text OCR (Screenshots, Scans, Photos)
Pricing
from $2.00 / 1,000 image reads
Image to Text OCR (Screenshots, Scans, Photos)
Extract text from images by URL: screenshots, scanned documents, receipts, photos of pages. Paragraph-aware text, confidence scores, optional word boxes. 7 languages.
Pricing
from $2.00 / 1,000 image reads
Rating
0.0
(0)
Developer
Scot C
Maintained by CommunityActor stats
0
Bookmarked
2
Total users
1
Monthly active users
20 hours ago
Last modified
Categories
Share
Extract text from images in bulk: screenshots, scanned documents, receipts, photos of pages, slides and UI captures. Paste image links and get clean, paragraph-aware text back, plus confidence scores so you know which results to trust.
Features
- Paragraph-aware text: lines are grouped into paragraphs, and words hyphenated across line breaks are re-joined.
- Confidence scores: average confidence per image plus a count of low-confidence words, so automations can flag poor scans.
- Automatic enhancement: EXIF auto-rotation (phone photos), grayscale, auto-contrast, and upscaling of small screenshots for better accuracy.
- Layout modes: Auto (documents and screenshots), Single block (a cropped paragraph) or Sparse (signs, labels, UI text).
- Word positions (optional): every word with its bounding box and confidence, in original-image pixels.
- 7 languages: English, German, French, Spanish, Italian, Portuguese and Dutch, plus mixed English/German, English/French and English/Spanish.
- Formats: PNG, JPG, WEBP, TIFF, BMP and GIF (first frame), up to 25 MB.
- Private by design: it runs open-source Tesseract inside the Actor. Your images aren't sent to any third-party OCR API.
Who it's for
- Automation builders (Make, Zapier, n8n): turn screenshots or uploaded images into text in one API call.
- AI pipelines: feed text from images into LLMs and RAG.
- Back-office work: read receipts, labels and scanned letters in bulk.
Input
| Field | Example |
|---|---|
| Image URLs | https://example.com/receipt.jpg |
| Language | English |
| Layout | Auto |
| Enhance image | on |
| Include word positions | off |
Output (one row per image)
Pricing
Pay per image successfully read. Failed downloads and unreadable files are never charged. The Actor also respects Apify's maximum-charge-per-run setting.
Good to know
- Works best on printed text. Handwriting and very blurry or low-resolution photos give weaker results; check
meanConfidence. - For PDF files, use PDF to Markdown for AI & RAG. It handles text-layer PDFs and runs OCR on scanned pages automatically.
- Only process images you have the right to use.
