Live Free tier included

Document OCR & Data Extraction

Extract text from images and scanned documents using Tesseract OCR. Auto-detects emails, phone numbers, dates, and monetary amounts with structured output.

Try it Free View Docs

Tesseract OCR Auto-extract Data Multi-language Receipts & Invoices

Interactive Playground

Try it Free

Upload an image and extract text with structured data in seconds.

Try Free

Upload Image

Language

Document Type

Uploading image...

Result

Upload an image to extract text

API Documentation

Extract text from images and scanned documents using Tesseract OCR. Auto-detects emails, phone numbers, dates, and monetary amounts. Supports 20+ languages including Arabic.

Endpoint

POSThttps://www.mahmoudalhabash.com/api/v1/ocr

Upload an image file or send base64-encoded image data. The API runs Tesseract OCR and returns extracted text with word-level confidence scores and structured data extraction.

Request Parameters

Parameter	Type	Required	Default	Description
`image`	file	required*	—	Image file (JPG, PNG, GIF, WebP, BMP, TIFF, max 10MB)
`base64`	string	required*	—	Base64-encoded image (alternative to file upload)
`language`	string	optional	eng	Tesseract language code (eng, ara, fra, deu, spa, etc.)
`type`	string	optional	auto	Document type: auto, receipt, invoice, id_card, general

* Either image file or base64 is required.

Response Fields

text Full extracted text

word_count Number of words extracted

confidence Average OCR confidence (0–100)

structured.emails Auto-detected email addresses

structured.phones Auto-detected phone numbers

structured.dates Auto-detected dates

structured.amounts Auto-detected monetary amounts

structured.total Detected total amount (receipts/invoices)

blocks Word-level bounding boxes with confidence scores

Code Examples

cURL (file upload)

curl -X POST https://www.mahmoudalhabash.com/api/v1/ocr \ -F "image=@receipt.jpg" \ -F "language=eng" \ -F "type=receipt"

Response 200JSON

{ "status": "success", "text": "Coffee Shop\nDate: 03/11/2026\n...", "word_count": 45, "confidence": 94.2, "structured": { "emails": [], "phones": ["+1-555-0123"], "dates": ["03/11/2026"], "amounts": ["$4.50", "$12.99"], "total": "12.99" }, "took_ms": 1200 }

Python (requests)

import requests with open('document.png', 'rb') as f: resp = requests.post( 'https://www.mahmoudalhabash.com/api/v1/ocr', files={'image': f}, data={ 'language': 'eng', 'type': 'auto' } ) data = resp.json() print(f"Text: {data['text'][:200]}") print(f"Confidence: {data['confidence']}%") print(f"Emails: {data['structured']['emails']}") print(f"Amounts: {data['structured']['amounts']}")

Error Codes

422 Validation failed — image or base64 required, max 10MB

422 No text extracted — try a clearer image or different language

429 Rate limit exceeded

500 Tesseract OCR processing failed

Rate Limits

Unlimited — Free during beta

Powered by Tesseract OCR v5. Max file size 10MB. Supports file upload or base64 input. No data is stored — images are processed and immediately discarded.