Optical Character Recognition (OCR)

Quick Answer: The Azure AI Vision Read API extracts printed text (164 languages) and handwritten text (9 languages) from images and PDFs. Results are organized as pages → lines → words with bounding polygon coordinates. For structured document extraction (invoices, forms), use Document Intelligence instead.

Read API vs. Document Intelligence

Using the Read API

Synchronous (Small Images)

from azure.ai.vision.imageanalysis import ImageAnalysisClient
from azure.ai.vision.imageanalysis.models import VisualFeatures
from azure.core.credentials import AzureKeyCredential

client = ImageAnalysisClient(
    endpoint="https://my-vision.cognitiveservices.azure.com/",
    credential=AzureKeyCredential("<your-key>")
)

# Read text from an image
with open("document.jpg", "rb") as f:
    result = client.analyze(
        image_data=f.read(),
        visual_features=[VisualFeatures.READ]
    )

# Extract text
for block in result.read.blocks:
    for line in block.lines:
        print(f"Line: {line.text}")
        print(f"  Bounding polygon: {line.bounding_polygon}")
        for word in line.words:
            print(f"  Word: {word.text} "
                  f"(confidence: {word.confidence:.2f})")

Read API Response Structure

{
    "readResult": {
        "blocks": [
            {
                "lines": [
                    {
                        "text": "Hello World",
                        "boundingPolygon": [
                            {"x": 10, "y": 10},
                            {"x": 200, "y": 10},
                            {"x": 200, "y": 40},
                            {"x": 10, "y": 40}
                        ],
                        "words": [
                            {
                                "text": "Hello",
                                "boundingPolygon": [...],
                                "confidence": 0.99
                            },
                            {
                                "text": "World",
                                "boundingPolygon": [...],
                                "confidence": 0.98
                            }
                        ]
                    }
                ]
            }
        ]
    }
}

Language Support

Best Practices for OCR Accuracy

On the Exam: Know when to use the Read API (general text extraction) vs. Document Intelligence (structured document processing). If the question mentions invoices, receipts, or forms — the answer is Document Intelligence. If it mentions signs, labels, or general images — the answer is the Read API.