다음을 통해 공유

Custom Vision

  • Reference

Azure Custom Vision Service is a Microsoft Cognitive Services product for tagging images using your custom computer vision model.

This connector is available in the following products and regions:

Service Class Regions
Logic Apps Standard All Logic Apps regions except the following:
     -   Azure China regions
Power Automate Standard All Power Automate regions except the following:
     -   China Cloud operated by 21Vianet
Power Apps Standard All Power Apps regions except the following:
     -   China Cloud operated by 21Vianet
Contact
Name Microsoft
URL Microsoft LogicApps Support
Microsoft Power Automate Support
Microsoft Power Apps Support
Connector Metadata
Publisher Microsoft
Website https://azure.microsoft.com/services/cognitive-services/custom-vision-service/

Creating a connection

The connector supports the following authentication types:
Default Parameters for creating connection. All regions Shareable

Default

Applicable: All regions

Parameters for creating connection.

This is shareable connection. If the power app is shared with another user, connection is shared as well. For more information, please see the Connectors overview for canvas apps - Power Apps | Microsoft Docs

Name Type Description Required
Prediction Key securestring Custom Vision Service Prediction Key True
Site URL string Root site url. If not specified, site url will be defaulted to https://southcentralus.api.cognitive.microsoft.com.

Throttling Limits

Name Calls Renewal Period
API calls per connection 1200 60 seconds

Actions

Classify an image (V2)

Classify an image based on image content (V2).
Classify an image [DEPRECATED]

This action has been deprecated. Please use Classify an image (V2) instead.

Classify an image based on image content.
Classify an image url (V2)

Classify an image located at the specified URL (V2).
Classify an image url [DEPRECATED]

This action has been deprecated. Please use Classify an image url (V2) instead.

Classify an image located at the specified URL.
Detect objects in an image (V2)

Detect objects in an image based on image content (V2).
Detect objects in an image [DEPRECATED]

This action has been deprecated. Please use Detect objects in an image (V2) instead.

Detect objects in an image based on image content.
Detect objects in an image url (V2)

Detect objects in an image located at the specified URL (V2).
Detect objects in an image url [DEPRECATED]

This action has been deprecated. Please use Detect objects in an image url (V2) instead.

Detect objects in an image located at the specified URL.
Predict tags from image [DEPRECATED]

Predict the tags for an image based on image content.
Predict tags from image URL [DEPRECATED]

Predict the tags for an image located at the specified URL.

Classify an image (V2)

Operation ID:
ClassifyImageV2

Classify an image based on image content (V2).

Parameters

Name Key Required Type Description
Project ID
 projectId True string

The ID of the Custom Vision Project.
Published Name
 publishedName True string

The name of the model to evaluate against.
Image Content
 Image True binary

The image you wish to analyze.

Returns

Body
PredictImage_ResponseV3

Classify an image [DEPRECATED]

Operation ID:
ClassifyImage

This action has been deprecated. Please use Classify an image (V2) instead.

Classify an image based on image content.

Parameters

Name Key Required Type Description
Project ID
 projectId True string

The ID of the Custom Vision Project.
Published Name
 publishedName True string

The name of the model to evaluate against.
Image Content
 Image True binary

The image you wish to analyze.

Returns

Body
PredictImage_ResponseV2

Classify an image url (V2)

Operation ID:
ClassifyImageUrlV2

Classify an image located at the specified URL (V2).

Parameters

Name Key Required Type Description
Project ID
 projectId True string

The ID of the Custom Vision Project.
Published Name
 publishedName True string

The name of the model to evaluate against.
Image URL
 Url True string

The URL of the image to analyze.

Returns

Body
PredictImage_ResponseV3

Classify an image url [DEPRECATED]

Operation ID:
ClassifyImageUrl

This action has been deprecated. Please use Classify an image url (V2) instead.

Classify an image located at the specified URL.

Parameters

Name Key Required Type Description
Project ID
 projectId True string

The ID of the Custom Vision Project.
Published Name
 publishedName True string

The name of the model to evaluate against.
Image URL
 Url True string

The URL of the image to analyze.

Returns

Body
PredictImage_ResponseV2

Detect objects in an image (V2)

Operation ID:
DetectImageV2

Detect objects in an image based on image content (V2).

Parameters

Name Key Required Type Description
Project ID
 projectId True string

The ID of the Custom Vision Project.
Published Name
 publishedName True string

The name of the model to evaluate against.
Image Content
 Image True binary

The image you wish to analyze.

Returns

Body
PredictImage_ResponseV3

Detect objects in an image [DEPRECATED]

Operation ID:
DetectImage

This action has been deprecated. Please use Detect objects in an image (V2) instead.

Detect objects in an image based on image content.

Parameters

Name Key Required Type Description
Project ID
 projectId True string

The ID of the Custom Vision Project.
Published Name
 publishedName True string

The name of the model to evaluate against.
Image Content
 Image True binary

The image you wish to analyze.

Returns

Body
PredictImage_ResponseV2

Detect objects in an image url (V2)

Operation ID:
DetectImageUrlV2

Detect objects in an image located at the specified URL (V2).

Parameters

Name Key Required Type Description
Project ID
 projectId True string

The ID of the Custom Vision Project.
Published Name
 publishedName True string

The name of the model to evaluate against.
Image URL
 Url True string

The URL of the image to analyze.

Returns

Body
PredictImage_ResponseV3

Detect objects in an image url [DEPRECATED]

Operation ID:
DetectImageUrl

This action has been deprecated. Please use Detect objects in an image url (V2) instead.

Detect objects in an image located at the specified URL.

Parameters

Name Key Required Type Description
Project ID
 projectId True string

The ID of the Custom Vision Project.
Published Name
 publishedName True string

The name of the model to evaluate against.
Image URL
 Url True string

The URL of the image to analyze.

Returns

Body
PredictImage_ResponseV2

Predict tags from image [DEPRECATED]

Operation ID:
PredictImage

Predict the tags for an image based on image content.

Parameters

Name Key Required Type Description
Project ID
 projectId True string

The ID of the Custom Vision Project.
Iteration ID
 iterationId string

The id of a particular iteration to evaluate against. The default iteration for the project will be used when not specified.
Image Content
 Image True binary

The image you wish to analyze.

Returns

Body
PredictImageURL_Response

Predict tags from image URL [DEPRECATED]

Operation ID:
PredictImageURL

Predict the tags for an image located at the specified URL.

Parameters

Name Key Required Type Description
Project ID
 projectId True string

The ID of the Custom Vision Project.
Iteration ID
 iterationId string

The id of a particular iteration to evaluate against. The default iteration for the project will be used when not specified.
Image URL
 Url True string

The URL of the image to analyze.

Returns

Body
PredictImageURL_Response

Definitions

PredictImageURL_Response

Name Path Type Description
Created
 Created date-time

The date-time the prediction was created.
Id
 Id string

The unique identifier of the prediction.
Iteration
 Iteration string

The iteration the prediction was evaluated against.
Predictions
 Predictions array of object

Predictions
Probability
 Predictions.Probability float

The probability of the prediction.
Tag
 Predictions.Tag string

The tag from the prediction.
TagId
 Predictions.TagId string

The unique identifier of the tag.
Project
 Project string

The project the prediction was evaluated against.

PredictImage_ResponseV2

Name Path Type Description
Id
 id string

The unique identifier of the prediction.
Project
 project string

The project the prediction was evaluated against.
Iteration
 iteration string

The iteration the prediction was evaluated against.
Created
 created date-time

The date-time the prediction was created.
Predictions
 Predictions array of object

Predictions
Probability
 Predictions.probability float

The probability of the prediction.
Tag
 Predictions.tagName string

The tag from the prediction.
TagId
 Predictions.tagId string

The unique identifier of the tag.
Left
 Predictions.boundingBox.left float

The x-coordinate of the left-top corner.
Top
 Predictions.boundingBox.top float

The y-coordinate of the left-top corner.
Width
 Predictions.boundingBox.width float

The width of the bounding box.
Height
 Predictions.boundingBox.height float

The height of the bounding box.

PredictImage_ResponseV3

Name Path Type Description
Id
 id string

The unique identifier of the prediction.
Project
 project string

The project the prediction was evaluated against.
Iteration
 iteration string

The iteration the prediction was evaluated against.
Created
 created date-time

The date-time the prediction was created.
Predictions
 predictions array of object

Predictions
Probability
 predictions.probability float

The probability of the prediction.
Tag
 predictions.tagName string

The tag from the prediction.
TagId
 predictions.tagId string

The unique identifier of the tag.
Left
 predictions.boundingBox.left float

The x-coordinate of the left-top corner.
Top
 predictions.boundingBox.top float

The y-coordinate of the left-top corner.
Width
 predictions.boundingBox.width float

The width of the bounding box.
Height
 predictions.boundingBox.height float

The height of the bounding box.