Issue with AI computer vision image analysis and OCR in vision studio portal part of Microsoft Foundry

Hello Somnath Das,

Welcome to Microsoft Q&A and Thank you for reaching out,

I understand why this is confusing, especially if you were previously using Image Analysis and OCR from the portal.

Let me clarify the current situation.

Current behavior in Vision Studio (Foundry)

What you’re seeing is expected.

In the Foundry “Vision Studio” experience, only the Face service is currently exposed in the UI

Image Analysis and OCR are not visible in the portal, which can give the impression that they’ve been removed

Are Image Analysis and OCR deprecated?

• These features are not removed today
• However, legacy versions are marked for deprecation with a retirement timeline (currently September 25, 2028)

So Services are still available and supported, They are not actively surfaced in the Foundry UI anymore

Why this change happened

• Foundry is evolving toward a streamlined and scenario-focused UI
• Some services are now:
  • Considered legacy within the Foundry portal experience
  • Still fully accessible via APIs and SDKs

How you can still use Image Analysis and OCR

1. Use Vision Studio

Go to:

https://portal.vision.cognitive.azure.com/

Here you can still test:

• Image Analysis
• OCR (Read API)

Make sure:

• You select the correct subscription and Vision resource

2.Use REST APIs / SDKs

You can continue using both services programmatically:

Image Analysis

Use Analyze Image API (v3.2 or v4.0)

Supports Tags, objects, captions, People detection and Smart crops

OCR

Use Read API

Supports Printed and handwritten text, Multi-page PDFs, Async processing for large documents

Example

.NET (Image Analysis v4.0 style usage):

var client = new ComputerVisionClient(new Uri(endpoint), new AzureKeyCredential(key));

Python (OCR Read API):

poller = client.begin_read(image_url)

Things to verify

Ensure you are using A valid Azure AI Vision / Cognitive Services resource

Confirm Correct region and Correct resource selected in portal

For learning Use Vision Studio portal (standalone)

For real-world usage Use REST API / SDK integration, as this is the most reliable and future-proof approach

Please refer this

What is Azure Vision in Foundry Tools? (deprecation notice) https://dori-uw-1.kuma-moon.com/azure/ai-services/computer-vision/overview?wt.mc_id=knowledgesearch_inproduct_azure-cxp-community-insider

Quickstart: Image Analysis client library v4.0 https://dori-uw-1.kuma-moon.com/azure/ai-services/computer-vision/quickstarts-sdk/image-analysis-client-library-40

Quickstart: OCR (Read API) https://dori-uw-1.kuma-moon.com/azure/ai-services/computer-vision/quickstarts-sdk/client-library?tabs=windows%2Cvisual-studio

I Hope this helps. Do let me know if you have any further queries.

If this answers your query, please do click Accept Answer and Yes for was this answer helpful.

Thank you!

Image Analysis and OCR capabilities in Azure Vision are not turned off, but the legacy experiences are being deprecated and replaced, which is why they no longer appear in Vision Studio the way they did before.

From the current documentation:

1. Image Analysis status
   • The Image Analysis service (version 4.0) is still available and generally available.
   • However, Image Analysis 4.0 in Azure Vision in Foundry Tools is deprecated and will be retired on September 25, 2028. After that date, calls to that service will fail.
   • Image Analysis 4.0 supports features such as captioning, dense captions, OCR, people detection, content moderation, smart crops, and face/person detection.
   • Image Analysis can be used via the client library SDKs or directly through the REST API, following the quickstarts and how-to guides.
2. OCR status
   • The legacy OCR APIs (v2.1, v3.2 Read) are not recommended anymore.
   • For OCR scenarios, Microsoft now recommends two main options:
     • OCR for images (version 4.0) as part of Azure Vision Image Analysis for general, in-the-wild images (labels, street signs, posters, etc.).
     • Azure Document Intelligence Read model for text-heavy scanned and digital documents (books, articles, reports, invoices, etc.).
3. Why features may be missing in Vision Studio
   • Because the legacy Image Analysis and OCR experiences are deprecated, the Vision Studio UI may no longer expose those older tiles or flows, even though the underlying capabilities remain available through:
     • Image Analysis 4.0 (for image understanding, OCR in images, content moderation, people/faces detection, smart crops, etc.).
     • Document Intelligence (for document-centric OCR).
4. How to use these capabilities now
   • For general image understanding and OCR in images:
     • Use the Image Analysis 4.0 APIs or SDKs as described in the Image Analysis quickstarts and how-to guides.
   • For document OCR (forms, invoices, reports, etc.):
     • Use Azure Document Intelligence and its Read model.
   • For face detection/recognition scenarios beyond simple detection:
     • Use the Azure AI Face service, which provides detection plus recognition (identification, verification, etc.).

In summary, Image Analysis and OCR are not fully removed as services, but the legacy Vision Studio experiences are deprecated. The recommended path is to use Image Analysis 4.0 for image-based scenarios and Azure Document Intelligence for document OCR, accessed via SDKs or REST APIs rather than relying solely on the old Vision Studio tiles.

References:

• What is Azure Vision in Foundry Tools?
• What is Image Analysis?
• OCR - Optical Character Recognition
• Migrate from Azure Vision in Foundry Tools - Image Analysis
• Transparency note: Image Analysis
• Microsoft Azure AI Fundamentals: Computer Vision