azure-ai-vision-imageanalysis-py - Azure AI Vision Image Analysis SDK for Python
Python SDK for Azure AI Vision 4.0 image analysis supporting captions, tags, object detection, OCR, people detection, and smart cropping.
Tags
Updated: 2026-09-17Capabilities
Typical Inputs
Typical Outputs
What this skill does
- Analyze images from URLs
- Analyze images from files
- Generate image captions
- Extract image tags
- Detect objects in images
- Extract text using OCR
- Detect people in images
- Generate smart crop regions
Inputs
- Azure Vision endpoint
- Azure Vision API key or Entra ID credentials
- Image URL or image file data
Outputs
- Image captions and confidence scores
- Image tags and confidence scores
- Detected object bounding boxes
- OCR text lines and words
- People detection bounding boxes
- Smart crop bounding boxes
Requirements
- Python package azure-ai-vision-imageanalysis
- Azure AI Vision resource endpoint and authentication
- Image format JPEG, PNG, GIF, BMP, WEBP, ICO, TIFF, or MPO
- Image size up to 20 MB
- Image dimensions between 50x50 and 16000x16000 pixels