Image OCR Extractor

Extract text from images (PNG, JPG, WebP, TIFF, BMP) using Tesseract OCR. Batch up to 50 images, multi-language (100+ languages), confidence scores. $0.002/image.

---

Why this exists

Images containing text — screenshots, scanned forms, product labels, street signs, whiteboards — can't be searched or processed until the text is extracted. This actor handles the OCR pipeline: download the image, preprocess it, run Tesseract, and return structured output with confidence scores.

---

What you get

{
  "index": 1,
  "url": "https://example.com/screenshot.png",
  "ok": true,
  "text": "Invoice #2024-001\nBill To: Acme Corp\nTotal: $1,250.00",
  "word_count": 9,
  "confidence": 87.4,
  "language": "eng",
  "error": null
}

---

Input Parameters

Parameter	Type	Default	Description
urls	array	—	List of image URLs (batch mode, up to 50)
url	string	—	Single image URL (used when urls is empty)
language	string	eng	Tesseract language code (eng, deu, fra, spa, jpn, chi_sim, kor, or combined eng+deu)
preprocessing	string	grayscale	Image preprocessing: none, grayscale, or threshold (binarize)
pageSegMode	integer	3	Tesseract PSM: 3=auto, 6=block, 7=line, 8=word, 11=sparse
limit	integer	50	Max images to process (cap: 200)

---

Quick start

Single image

{
  "url": "https://example.com/invoice.png"
}

Batch of screenshots

{
  "urls": [
    "https://example.com/slide1.png",
    "https://example.com/slide2.png"
  ],
  "language": "eng",
  "preprocessing": "grayscale"
}

Japanese document

{
  "url": "https://example.com/japanese-receipt.jpg",
  "language": "jpn",
  "preprocessing": "threshold",
  "pageSegMode": 6
}

---

Preprocessing Tips

Mode	Best for
grayscale	Photos, colored backgrounds, most general use
threshold	Documents with dark text on white background, typed forms
none	When the image is already clean B&W

---

Supported Languages (pre-installed)

eng (English), deu (German), fra (French), spa (Spanish), jpn (Japanese), chi_sim (Simplified Chinese), kor (Korean)

Multi-language: "eng+fra" — useful for bilingual documents.

---

Pricing

Pay-Per-Event: $0.002 per image successfully OCR-processed.

Images	Cost
10 screenshots	$0.02
50 images (full batch)	$0.10
200 product photos	$0.40

---

Use cases

1. Invoice / receipt digitization — Extract line items, totals, vendor names
2. Screenshot text extraction — Turn UI screenshots into searchable text
3. Multilingual document processing — Japanese, Chinese, Korean, European languages
4. Form digitization — Handwritten or typed form fields
5. Whiteboard photos — Meeting notes, diagrams with text

---

Confidence score

Each result includes an confidence score (0–100) from Tesseract:

• 85–100: High quality OCR, likely accurate
• 60–84: Moderate, review critical fields
• < 60: Low confidence — consider using threshold preprocessing or higher-resolution source image

---

Related actors (same author)

• PDF Text & Table Extractor — PDF version with table detection
• Web Page → Markdown Converter — HTML text extraction

---

Feedback

Leave a review on Apify Store