Object Detection Activity
The Object Detection activity enables the identification of objects and living things within an image by returning their specific bounding box coordinates in pixels. This functionality allows users to analyze spatial relationships between items and apply descriptive tags based on the identified visual content.
| Field | Description | Requirement |
|---|---|---|
| Ocp Apim Subscription Key | The unique Microsoft Computer Vision API key used for service authentication. | Required |
| Endpoint | The specific Microsoft Computer Vision API URL associated with the service instance. | Required |
| Image Url | The publicly accessible URL address of the image file to be processed. | Required |
| Response variable name | The name of the variable designated to store the HTTP response data. | Required |
Action Types & Examples
Object Identification
- Format: string
- Example Result: "{"objects":[{"rectangle":{"x":6,"y":390,"w":48,"h":40},"object":"person"}]}"
Implementation Examples
Field Setup - Ocp Apim Subscription Key: "2dfc034684a3434fb3270534c5efdc66" - Endpoint: "https://yourapiurl.cognitiveservices.azure.com/" - Image Url: "http://example.com/images/test.jpg"
Execution Parameters - Response variable name: "getResponse"
Technical Notes
This activity integrates with Microsoft Cognitive Services. For successful execution, ensure the image URL is publicly reachable and the API subscription key is active with sufficient quota. Detailed technical specifications regarding the underlying computer vision features can be found in the official Microsoft Azure documentation.