Actions21
- Azure OpenAI Actions
- Batch API Actions
- Content Safety Actions
- Document Intelligence Actions
- Language Actions
- Speech Actions
- Translator Actions
- Vision Actions
Vision → Analyze Image
AI-generatedSummary
Analyze an image with Azure AI Vision, selecting one or more features to return captions, tags, detected objects, OCR text, people, or smart crop regions. Provide either a publicly accessible image URL or the name of an input binary property containing the image.
Inputs
- Authentication (required) — Choose API Key or Entra ID (OAuth2) authentication and configure the matching Azure credential.
- Image Source (required) — Choose URL or Binary File. For URL, provide the publicly accessible Image URL. For Binary File, set Binary Property to the name of the input item's binary field; it defaults to data.
- Features (required) — Select at least one feature to run: Caption, Tags, Objects, Read (OCR), People, or Smart Crops. Defaults to Caption and Tags.
Output shape
One JSON result per input item, containing fields for the selected analyses and the raw Azure response in _raw.
The shaped result can include caption, tags, objects, text, people, and smartCrops; fields are included when Azure returns the corresponding result. The node sends the image to Azure AI Vision for analysis.