Image OCR — Extract Text from Images

Pricing

Rating

Developer

Actor stats

Categories

Share

Turn images into text. Send one or more image URLs and get back the recognized text with a confidence score — powered by the Tesseract OCR engine, with support for 100+ languages.

Perfect for receipts, invoices, screenshots, photos of documents, business cards, signs, labels and any picture that contains text.

Features

• 🔤 Accurate OCR — built on Tesseract, a battle-tested open-source OCR engine.
• 🌍 Multi-language — English, Italian, French, German, Spanish, Portuguese and many more; combine several for multilingual images.
• 📈 Confidence score — every result includes how confident the engine is.
• 📦 Batch — extract text from many images in a single run.
• 🛡️ Robust — no website to scrape, nothing to break: pure on-engine processing.

Input

Example input

{
"imageUrls":[
"https://tesseract.projectnaptha.com/img/eng_bw.png"
],
"languages":["eng"]
}

Output

One dataset item per image:

{
"url":"https://tesseract.projectnaptha.com/img/eng_bw.png",
"success":true,
"language":"eng",
"confidence":92.4,
"text":"The (quick) [brown] {fox} jumps! Over the $43,456.78 ..."
}

Export as JSON, CSV, or Excel, or pull via the Apify API.

Common use cases

• Digitize receipts and invoices into searchable text.
• Extract text from screenshots and photos of documents.
• Read business cards, labels, signs and forms.
• Feed image text into search, spreadsheets, or LLM/RAG pipelines.

Notes

• Works best on clear, high-contrast images. Blurry or low-resolution photos reduce accuracy.
• For multilingual images, pass all relevant language codes (e.g. ["eng","ita"]).
• imageUrls must be direct links to the image file.