About Microsoft Azure Computer Vision
Microsoft Azure Computer Vision provides businesses with advanced AI-powered tools to analyze and interpret visual content such as images and videos. It offers capabilities including optical character recognition (OCR) to extract printed and handwritten text from various documents, object detection to identify and classify thousands of items, and facial analysis for identity verification and customer insights. The service supports multiple languages and works across diverse business scenarios like retail inventory management, logistics package verification, and security operations. By automating visual data extraction and analysis, it enables companies to streamline workflows, improve operational efficiency, and gain deeper understanding of their visual assets. Integration with Azure Cognitive Services allows seamless embedding of these capabilities into business applications without requiring deep AI expertise.
Available Actions
These are the specific actions that AI agents can perform with this tool
Analyze an image to extract visual features including categories, tags, and description.
url
The URL of the image to analyze
features
Comma-separated list of visual features to analyze
language
The language of the returned metadata
Generate Thumbnail
7 inputs
Generate a thumbnail from a given image URL with specified dimensions using Azure AI.
Endpoint
Supported Cognitive Services endpoint.
height
Height of the thumbnail in pixels.
width
Width of the thumbnail in pixels.
model-version
AI model version, defaults to 'latest'.
smartCropping
Enable smart cropping.
Ocp-Apim-Subscription-Key
API key for authentication
url
Publicly reachable URL of an image.
Retrieve the list of domain-specific models supported by the Computer Vision API.
Endpoint
Supported Cognitive Services endpoints.
Ocp-Apim-Subscription-Key
The API key used for authenticating requests.
Recognize Printed Text
6 inputs
Use the Azure OCR API to recognize and extract text from an image.
Endpoint
Supported Cognitive Services endpoint.
detectOrientation
Whether to detect the text orientation in the image.
language
Language code of the text in the image. Defaults to 'unk'.
model-version
Version of the AI model. Defaults to 'latest'.
Ocp-Apim-Subscription-Key
Subscription key for accessing the service.
url
Publicly reachable URL of the image.
The endpoint tags an image with relevant words related to its content.
Endpoint
Supported Cognitive Services endpoints.
url
Publicly reachable URL of an image.
language
The desired language for output generation. Defaults to 'en'.
modelVersion
Optional parameter to specify the version of the AI model. Defaults to 'latest'.