MSDN Documentation

Featured Tutorials

• Getting Started with Image Analysis

  Learn how to use the Vision API to detect objects, describe scenes, identify celebrities, and analyze adult content.

  Learn More →

• Optical Character Recognition (OCR)

  Extract text from images and documents to make your content searchable and editable.

  Learn More →

• Face Detection and Recognition

  Understand how to detect faces, analyze attributes, and group similar faces using the Face API.

  Learn More →

• Custom Vision Service: Image Classification

  Build your own custom image classification models tailored to your specific needs.

  Learn More →

• Custom Vision Service: Object Detection

  Train models to detect and locate specific objects within images.

  Learn More →

• Video Indexer: Extract Insights from Videos

  Discover how to analyze video content for insights like spoken words, faces, sentiments, and more.

  Learn More →

Key Concepts

The Vision API offers a range of powerful capabilities:

• Image Analysis: Provides rich metadata about the content of an image, including descriptions, tags, objects, and brands.
• OCR: Enables you to read text from images and documents.
• Face API: Detects and analyzes human faces, including attributes like age, gender, emotion, and identification.
• Custom Vision: Allows you to build and deploy custom machine learning models for image classification and object detection with your own data.
• Video Indexer: A comprehensive video analytics service that extracts insights from video content.

Prerequisites

To follow these tutorials, you'll generally need:

• An Azure account (free trial available).
• A Vision API resource created in Azure.
• Appropriate SDKs or REST API tools installed on your development machine (e.g., Python, C#, Node.js).

Code Examples

Many tutorials include code snippets to help you implement these features quickly. Here's a basic example of calling the Image Analysis API using Python:

import requests
import os

# Replace with your Vision API subscription key and endpoint
subscription_key = os.environ.get('VISION_KEY')
endpoint = os.environ.get('VISION_ENDPOINT')

analyze_url = endpoint + "vision/v3.2/analyze"

image_url = "https://www.microsoft.com/en-us/research/wp-content/uploads/2016/02/bigdata-1480x987.jpg"

headers = {
    'Ocp-Apim-Subscription-Key': subscription_key,
    'Content-Type': 'application/json'
}

params = {
    'visualFeatures': 'Categories,Description,Tags,Objects',
    'language': 'en'
}

data = {
    'url': image_url
}

response = requests.post(analyze_url, headers=headers, params=params, json=data)
response.raise_for_status() # Raise an exception for bad status codes

analysis = response.json()

print("Description:", analysis['description']['captions'][0]['text'])
print("Tags:", [tag['name'] for tag in analysis['tags']])