[ ABORT TO HUD ]
SEQ. 1
SEQ. 2
SEQ. 3
SEQ. 4
SEQ. 5
Vision & Image Analysis
Computer Vision in Foundry
Foundry Tools (formerly Azure Cognitive Services) provide pre-built AI capabilities that you can plug into your applications via APIs.
Vision Capabilities
| Feature | What It Does | Use Case |
|---|---|---|
| Image Analysis 4.0 | Detect objects, read text (OCR), generate captions | Product cataloging, accessibility |
| Custom Vision | Train custom image classifiers | Defect detection, brand recognition |
| Face API | Detect and verify faces | Identity verification (with compliance) |
| Video Analysis | Extract insights from video content | Content moderation, scene detection |
Image Analysis Quick Start
from azure.ai.vision.imageanalysis import ImageAnalysisClient
from azure.identity import DefaultAzureCredential
client = ImageAnalysisClient(
endpoint="<your-foundry-endpoint>",
credential=DefaultAzureCredential()
)
result = client.analyze(
image_url="https://example.com/photo.jpg",
visual_features=["CAPTION", "OBJECTS", "READ"]
)
print(result.caption.text) # "A dog playing in a park"
💡 Key Insight: Vision APIs can be used as tools for AI agents. An agent can call the Vision API to understand images uploaded by users, enabling multimodal workflows within Foundry.
⌨ HANDS-ON LABAnalyze an Image over REST
⭐ +150 XPProduct photos need captions and OCR at scale. Grab your resource key with the CLI, then hit the Image Analysis 4.0 endpoint with curl.
1Fetch your Foundry resource's API keys with the Azure CLI.
2POST an image URL to imageanalysis:analyze requesting caption and read (OCR) features.
OBJECTIVE 1 / 2 — type "hint" if stuck
FOUNDRY VERIFICATION
QUERY 1 // 1
What does Image Analysis 4.0 provide?
Only face detection
Object detection, OCR text reading, and caption generation
Only image generation
Video editing