Computer Vision API

Extract rich information from images to categorize and process visual data—and protect your users from unwanted content with this Azure Cognitive Service.

Status: Production

Tier: Standard

Version: 1.0.0

Actions:

Name	Summary
ListModels ()	List domain specific models
GetImageSchema (string format)	Get Request Body Parameters
RecognizeDomainSpecificContent (string model, string format, RecognizeDomainSpecificContentParameterImage Image)	Recognize Domain Specific Content
TagImage (string format, TagImageParameterImage Image)	Tag Image
OCR ([internal][Optional]string language, [internal][Optional]boolean detectOrientation, string format, OCRParameterImage Image)	Optical Character Recognition (OCR) to JSON
OCRText ([internal][Optional]string language, [internal][Optional]boolean detectOrientation, string format, OCRTextParameterImage Image)	Optical Character Recognition (OCR) to Text
AnalyzeImage ([advanced][Optional]string language, [internal][Optional]string visualFeatures, string format, AnalyzeImageParameterImage Image)	Analyze Image
GetThumbnail (number width, number height, [advanced][Optional]boolean smartCropping, string format, GetThumbnailParameterImage Image)	Generate Thumbnail
DescribeImage ([advanced][Optional]number maxCandidates, string format, DescribeImageParameterImage Image)	Describe Image
DescribeImageContent ([advanced][Optional]number maxCandidates, string Image)	Describe Image Content
DescribeImageURL ([advanced][Optional]number maxCandidates, ImageContentObject ImageURL)	Describe Image URL
ListModelsV2 ()	List domain specific models
GetImageSchemaV2 (string format)	Get Request Body Parameters
RecognizeDomainSpecificContentV2 (string model, string format, RecognizeDomainSpecificContentV2ParameterImage Image)	Recognize Domain Specific Content
TagImageV2 (string format, TagImageV2ParameterImage Image)	Tag Image
OCRV2 ([internal][Optional]string language, [internal][Optional]boolean detectOrientation, string format, OCRV2ParameterImage Image)	Optical Character Recognition (OCR) to JSON
OCRTextV2 ([internal][Optional]string language, [internal][Optional]boolean detectOrientation, string format, OCRTextV2ParameterImage Image)	Optical Character Recognition (OCR) to Text
AnalyzeImageV2 ([advanced][Optional]string language, [internal][Optional]string visualFeatures, string format, AnalyzeImageV2ParameterImage Image)	Analyze Image
GetThumbnailV2 (number width, number height, [advanced][Optional]boolean smartCropping, string format, GetThumbnailV2ParameterImage Image)	Generate Thumbnail
DescribeImageV2 ([advanced][Optional]number maxCandidates, [advanced][Optional]string language, string format, DescribeImageV2ParameterImage Image)	Describe Image
DescribeImageContentV2 ([advanced][Optional]number maxCandidates, [advanced][Optional]string language, string Image)	Describe Image Content
DescribeImageURLV2 ([advanced][Optional]number maxCandidates, [advanced][Optional]string language, ImageContentObject ImageURL)	Describe Image URL
DetectObjects (string format, DetectObjectsParameterImage Image)	Detect Objects
GetAreaOfInterest (string format, GetAreaOfInterestParameterImage Image)	Get Area of Interest

Triggers:

Name

Summary

Objects:

Name	Summary
AnalyzeImageParameterImage
AnalyzeImageV2ParameterImage
AnalyzeResponse
AreaOfInterestResponse
DescribeImageParameterImage
DescribeImageV2ParameterImage
DescribeResponse
DetectObjectsParameterImage
DetectResponse
DomainModelResponse
GetAreaOfInterestParameterImage
GetImageSchemaResponse
GetImageSchemaV2Response
GetThumbnailParameterImage
GetThumbnailV2ParameterImage
ImageContentObject
ListModelsResponse
ModelResponse
OCRJsonResponse	JSON Response
OCRParameterImage
OCRTextParameterImage
OCRTextResponse
OCRTextV2ParameterImage
OCRV2ParameterImage
RecognizeDomainSpecificContentParameterImage
RecognizeDomainSpecificContentV2ParameterImage
TagImageParameterImage
TagImageV2ParameterImage
TagResponse

Actions:

ListModels

Summary: List domain specific models

Description: List domain specific models.

Syntax:

ComputerVisionAPI.ListModels ()

Returns:

Type:ListModelsResponse

GetImageSchema

Summary: Get Request Body Parameters

Description: Return the schema for the request body format desired.

Syntax:

ComputerVisionAPI.GetImageSchema (string format)

Parameters:

Name

Type

Summary

Required

Related Action

format

string

(Image Source)Values: [Image Content, Image URL]

Source of the image - either included or by reference url.

True

Returns:

Type:GetImageSchemaResponse

RecognizeDomainSpecificContent

Summary: Recognize Domain Specific Content

Description: Recognize celebrities or landmarks in an image.

Syntax:

ComputerVisionAPI.RecognizeDomainSpecificContent (string model, string format, RecognizeDomainSpecificContentParameterImage Image)

Parameters:

Name	Type	Summary	Required	Related Action
model	string (Domain Model)Values: [celebrities, landmarks]	Supported domain-specific model to recognize in image.	True
format	string (Image Source)Values: [Image Content, Image URL]	Source of the image - either included or by reference url.	True
Image	RecognizeDomainSpecificContentParameterImage		True

Returns:

Type:DomainModelResponse

TagImage

Summary: Tag Image

Description: Generates a list of words, or tags, that are relevant to the content of the supplied image.

Syntax:

ComputerVisionAPI.TagImage (string format, TagImageParameterImage Image)

Parameters:

Name

Type

Summary

Required

Related Action

format

string

(Image Source)Values: [Image Content, Image URL]

Source of the image - either included or by reference url.

True

Image

TagImageParameterImage

True

Returns:

Type:TagResponse

OCR

Summary: Optical Character Recognition (OCR) to JSON

Description: Optical Character Recognition (OCR) detects text in an image and extracts the recognized characters into a machine-usable JSON stream.

Syntax:

ComputerVisionAPI.OCR ([internal][Optional]string language, [internal][Optional]boolean detectOrientation, string format, OCRParameterImage Image)

Parameters:

Name	Type	Summary	Required	Related Action
language	string (Language)	The language code of the text to be detected in the image.	False
detectOrientation	boolean (Detect Orientation)	he BCP-47 language code of the text to be detected in the image.	False
format	string (Image Source)Values: [Image Content, Image URL]	Source of the image - either included or by reference url.	True
Image	OCRParameterImage		True

Returns:

Type:OCRJsonResponse

Summary: JSON Response

OCRText

Summary: Optical Character Recognition (OCR) to Text

Description: Optical Character Recognition (OCR) detects text in an image and extracts the recognized characters into a text file.

Syntax:

ComputerVisionAPI.OCRText ([internal][Optional]string language, [internal][Optional]boolean detectOrientation, string format, OCRTextParameterImage Image)

Parameters:

Name	Type	Summary	Required	Related Action
language	string (Language)	The language code of the text to be detected in the image.	False
detectOrientation	boolean (Detect Orientation)	The BCP-47 language code of the text to be detected in the image.	False
format	string (Image Source)Values: [Image Content, Image URL]	Source of the image - either included or by reference url.	True
Image	OCRTextParameterImage		True

Returns:

Type:OCRTextResponse

AnalyzeImage

Summary: Analyze Image

Description: This operation extracts a rich set of visual features based on the image content.

Syntax:

ComputerVisionAPI.AnalyzeImage ([advanced][Optional]string language, [internal][Optional]string visualFeatures, string format, AnalyzeImageParameterImage Image)

Parameters:

Name	Type	Summary	Required	Related Action
language	string (Language)Values: [en, zh]	The service will return recognition results in specified language.	False
visualFeatures	string (Visual Feature Options)	A string indicating what visual feature types to return. Multiple values should be comma-separated.	False
format	string (Image Source)Values: [Image Content, Image URL]	Source of the image - either included or by reference url.	True
Image	AnalyzeImageParameterImage		True

Returns:

Type:AnalyzeResponse

GetThumbnail

Summary: Generate Thumbnail

Description: This operation generates a thumbnail image with the user-specified width and height.

Syntax:

ComputerVisionAPI.GetThumbnail (number width, number height, [advanced][Optional]boolean smartCropping, string format, GetThumbnailParameterImage Image)

Parameters:

Name	Type	Summary	Required	Related Action
width	number (Thumbnail Width)	Width of the generated thumbnail - recommended is 50	True
height	number (Thumbnail Height)	Height of the generated thumbnail - recommended is 50	True
smartCropping	boolean (Smart Cropping)	Boolean flag for enabling smart cropping	False
format	string (Image Source)Values: [Image Content, Image URL]	Source of the image - either included or by reference url.	True
Image	GetThumbnailParameterImage		True

Returns:

Type:string(binary)

Summary: Thumbnail

Description: Generated thumbnail image

DescribeImage

Summary: Describe Image

Description: This operation generates a description of an image in human readable language with complete sentences.

Syntax:

ComputerVisionAPI.DescribeImage ([advanced][Optional]number maxCandidates, string format, DescribeImageParameterImage Image)

Parameters:

Name	Type	Summary	Required	Related Action
maxCandidates	number (Max Candidates)	Maximum number of candidate descriptions to be returned.	False
format	string (Image Source)Values: [Image Content, Image URL]	Source of the image - either included or by reference url.	True
Image	DescribeImageParameterImage		True

Returns:

Type:DescribeResponse

DescribeImageContent

Summary: Describe Image Content

Description: This operation generates a description of image content in human readable language with complete sentences.

Syntax:

ComputerVisionAPI.DescribeImageContent ([advanced][Optional]number maxCandidates, string Image)

Parameters:

Name

Type

Summary

Required

Related Action

maxCandidates

number

(Max Candidates)

Maximum number of candidate descriptions to be returned.

False

Image

string(binary)

True

Returns:

Type:DescribeResponse

DescribeImageURL

Summary: Describe Image URL

Description: This operation generates a description of an image URL in human readable language with complete sentences.

Syntax:

ComputerVisionAPI.DescribeImageURL ([advanced][Optional]number maxCandidates, ImageContentObject ImageURL)

Parameters:

Name

Type

Summary

Required

Related Action

maxCandidates

number

(Max Candidates)

Maximum number of candidate descriptions to be returned.

False

ImageURL

ImageContentObject

True

Returns:

Type:DescribeResponse

ListModelsV2

Summary: List domain specific models

Description: List domain specific models.

Syntax:

ComputerVisionAPI.ListModelsV2 ()

Returns:

Type:ListModelsResponse

GetImageSchemaV2

Summary: Get Request Body Parameters

Description: Return the schema for the request body format desired.

Syntax:

ComputerVisionAPI.GetImageSchemaV2 (string format)

Parameters:

Name

Type

Summary

Required

Related Action

format

string

(Image Source)Values: [Image Content, Image URL]

Source of the image - either included or by reference url.

True

Returns:

Type:GetImageSchemaV2Response

RecognizeDomainSpecificContentV2

Summary: Recognize Domain Specific Content

Description: Recognize celebrities or landmarks in an image.

Syntax:

ComputerVisionAPI.RecognizeDomainSpecificContentV2 (string model, string format, RecognizeDomainSpecificContentV2ParameterImage Image)

Parameters:

Name	Type	Summary	Required	Related Action
model	string (Domain Model)Values: [celebrities, landmarks]	Supported domain-specific model to recognize in image.	True
format	string (Image Source)Values: [Image Content, Image URL]	Source of the image - either included or by reference url.	True
Image	RecognizeDomainSpecificContentV2ParameterImage		True

Returns:

Type:DomainModelResponse

TagImageV2

Summary: Tag Image

Description: Generates a list of words, or tags, that are relevant to the content of the supplied image.

Syntax:

ComputerVisionAPI.TagImageV2 (string format, TagImageV2ParameterImage Image)

Parameters:

Name

Type

Summary

Required

Related Action

format

string

(Image Source)Values: [Image Content, Image URL]

Source of the image - either included or by reference url.

True

Image

TagImageV2ParameterImage

True

Returns:

Type:TagResponse

OCRV2

Summary: Optical Character Recognition (OCR) to JSON

Description: Optical Character Recognition (OCR) detects text in an image and extracts the recognized characters into a machine-usable JSON stream.

Syntax:

ComputerVisionAPI.OCRV2 ([internal][Optional]string language, [internal][Optional]boolean detectOrientation, string format, OCRV2ParameterImage Image)

Parameters:

Name	Type	Summary	Required	Related Action
language	string (Language)	The language code of the text to be detected in the image.	False
detectOrientation	boolean (Detect Orientation)	he BCP-47 language code of the text to be detected in the image.	False
format	string (Image Source)Values: [Image Content, Image URL]	Source of the image - either included or by reference url.	True
Image	OCRV2ParameterImage		True

Returns:

Type:OCRJsonResponse

Summary: JSON Response

OCRTextV2

Summary: Optical Character Recognition (OCR) to Text

Description: Optical Character Recognition (OCR) detects text in an image and extracts the recognized characters into a text file.

Syntax:

ComputerVisionAPI.OCRTextV2 ([internal][Optional]string language, [internal][Optional]boolean detectOrientation, string format, OCRTextV2ParameterImage Image)

Parameters:

Name	Type	Summary	Required	Related Action
language	string (Language)	The language code of the text to be detected in the image.	False
detectOrientation	boolean (Detect Orientation)	he BCP-47 language code of the text to be detected in the image.	False
format	string (Image Source)Values: [Image Content, Image URL]	Source of the image - either included or by reference url.	True
Image	OCRTextV2ParameterImage		True

Returns:

Type:OCRTextResponse

AnalyzeImageV2

Summary: Analyze Image

Description: This operation extracts a rich set of visual features based on the image content.

Syntax:

ComputerVisionAPI.AnalyzeImageV2 ([advanced][Optional]string language, [internal][Optional]string visualFeatures, string format, AnalyzeImageV2ParameterImage Image)

Parameters:

Name	Type	Summary	Required	Related Action
language	string (Language)Values: [en, ja, pt, zh]	The service will return recognition results in specified language.	False
visualFeatures	string (Visual Feature Options)	A string indicating what visual feature types to return. Multiple values should be comma-separated.	False
format	string (Image Source)Values: [Image Content, Image URL]	Source of the image - either included or by reference url.	True
Image	AnalyzeImageV2ParameterImage		True

Returns:

Type:AnalyzeResponse

GetThumbnailV2

Summary: Generate Thumbnail

Description: This operation generates a thumbnail image with the user-specified width and height.

Syntax:

ComputerVisionAPI.GetThumbnailV2 (number width, number height, [advanced][Optional]boolean smartCropping, string format, GetThumbnailV2ParameterImage Image)

Parameters:

Name	Type	Summary	Required	Related Action
width	number (Thumbnail Width)	Width of the generated thumbnail - recommended is 50	True
height	number (Thumbnail Height)	Height of the generated thumbnail - recommended is 50	True
smartCropping	boolean (Smart Cropping)	Boolean flag for enabling smart cropping	False
format	string (Image Source)Values: [Image Content, Image URL]	Source of the image - either included or by reference url.	True
Image	GetThumbnailV2ParameterImage		True

Returns:

Type:string(binary)

Summary: Thumbnail

Description: Generated thumbnail image

DescribeImageV2

Summary: Describe Image

Description: This operation generates a description of an image in human readable language with complete sentences.

Syntax:

ComputerVisionAPI.DescribeImageV2 ([advanced][Optional]number maxCandidates, [advanced][Optional]string language, string format, DescribeImageV2ParameterImage Image)

Parameters:

Name	Type	Summary	Required	Related Action
maxCandidates	number (Max Candidates)	Maximum number of candidate descriptions to be returned.	False
language	string (Language)Values: [en, ja, pt, zh]	The service will return recognition results in specified language.	False
format	string (Image Source)Values: [Image Content, Image URL]	Source of the image - either included or by reference url.	True
Image	DescribeImageV2ParameterImage		True

Returns:

Type:DescribeResponse

DescribeImageContentV2

Summary: Describe Image Content

Description: This operation generates a description of image content in human readable language with complete sentences.

Syntax:

ComputerVisionAPI.DescribeImageContentV2 ([advanced][Optional]number maxCandidates, [advanced][Optional]string language, string Image)

Parameters:

Name	Type	Summary	Required	Related Action
maxCandidates	number (Max Candidates)	Maximum number of candidate descriptions to be returned.	False
language	string (Language)Values: [en, ja, pt, zh]	The service will return recognition results in specified language.	False
Image	string(binary)		True

Returns:

Type:DescribeResponse

DescribeImageURLV2

Summary: Describe Image URL

Description: This operation generates a description of an image URL in human readable language with complete sentences.

Syntax:

ComputerVisionAPI.DescribeImageURLV2 ([advanced][Optional]number maxCandidates, [advanced][Optional]string language, ImageContentObject ImageURL)

Parameters:

Name	Type	Summary	Required	Related Action
maxCandidates	number (Max Candidates)	Maximum number of candidate descriptions to be returned.	False
language	string (Language)Values: [en, ja, pt, zh]	The service will return recognition results in specified language.	False
ImageURL	ImageContentObject		True

Returns:

Type:DescribeResponse

DetectObjects

Summary: Detect Objects

Description: Generates a list of deteced objects in the supplied image.

Syntax:

ComputerVisionAPI.DetectObjects (string format, DetectObjectsParameterImage Image)

Parameters:

Name

Type

Summary

Required

Related Action

format

string

(Image Source)Values: [Image Content, Image URL]

Source of the image - either included or by reference url.

True

Image

DetectObjectsParameterImage

True

Returns:

Type:DetectResponse

GetAreaOfInterest

Summary: Get Area of Interest

Description: Returns a bounding box around the most important area of the image.

Syntax:

ComputerVisionAPI.GetAreaOfInterest (string format, GetAreaOfInterestParameterImage Image)

Parameters:

Name

Type

Summary

Required

Related Action

format

string

(Image Source)Values: [Image Content, Image URL]

Source of the image - either included or by reference url.

True

Image

GetAreaOfInterestParameterImage

True

Returns:

Type:AreaOfInterestResponse

AnalyzeImageParameterImage

Summary:

Description:

Properties:

Name

Type

Summary

AnalyzeImageV2ParameterImage

Summary:

Description:

Properties:

Name

Type

Summary

AnalyzeResponse

Summary:

Description:

Properties:

Name	Type	Summary
categories	array of (CategoriesItem)	Categories identified from the image
description	Description	Properties to describe the image analyzed
tags	array of (TagsItem)	Tags identified with confidence scores.

CategoriesItem

Summary:

Description:

Properties:

Name

Type

Summary

name

string

Category Name

Name of the category identified from the image

score

number(float)

Category Confidence Score

Confidence Score for the image category

Description

Summary:

Description: Properties to describe the image analyzed

Properties:

Name

Type

Summary

captions

array of (CaptionsItem)

Captions

List of captions generated from the image

CaptionsItem

Summary:

Description:

Properties:

Name

Type

Summary

confidence

number(float)

Caption Confidence Score

Confidence score of the image caption

text

string

Caption Text

Text caption generated from the image

TagsItem

Summary:

Description:

Properties:

Name

Type

Summary

confidence

number(float)

Tag Confidence Score

Confidence score for the identified tags.

name

string

Tag Name

Name of the tag identified.

AreaOfInterestResponse

Summary:

Description:

Properties:

Name

Type

Summary

areaOfInterest

AreaOfInterest

Bounding box

AreaOfInterest

Summary:

Description: Bounding box

Properties:

Name

Type

Summary

DescribeImageParameterImage

Summary:

Description:

Properties:

Name

Type

Summary

DescribeImageV2ParameterImage

Summary:

Description:

Properties:

Name

Type

Summary

DescribeResponse

Summary:

Description:

Properties:

Name

Type

Summary

description

Description

Properties to describe the image analyzed

Description

Summary:

Description: Properties to describe the image analyzed

Properties:

Name

Type

Summary

captions

array of (CaptionsItem)

Captions

List of captions generated from the image

CaptionsItem

Summary:

Description:

Properties:

Name

Type

Summary

confidence

number(float)

Caption Confidence Score

Confidence score of the image caption

text

string

Caption Text

Text caption generated from the image

DetectObjectsParameterImage

Summary:

Description:

Properties:

Name

Type

Summary

DetectResponse

Summary:

Description:

Properties:

Name

Type

Summary

objects

array of (ObjectsItem)

Set of objects returned from the picture analysis.

ObjectsItem

Summary:

Description:

Properties:

Name	Type	Summary
rectangle	Rectangle Bounding box	Bounding box around the object
confidence	number(float) Object confidence score	Confidence score of the identified object.
object	string Object name	Name of the object identified.

Rectangle

Summary: Bounding box

Description: Bounding box around the object

Properties:

Name

Type

Summary

DomainModelResponse

Summary:

Description:

Properties:

Name

Type

Summary

result

Result

Computer Vision Result

Result

Summary:

Description: Computer Vision Result

Properties:

Name

Type

Summary

celebrities

array of (CelebritiesItem)

Recognized celebrities in image

landmarks

array of (LandmarksItem)

Recognized landmarks in image

CelebritiesItem

Summary:

Description:

Properties:

Name

Type

Summary

confidence

number(float)

Celebrity Confidence

Confidence score that image is of celebrity

name

string

Celebrity Name

Name of the recognized celebrity

LandmarksItem

Summary:

Description:

Properties:

Name

Type

Summary

confidence

number(float)

Landmark Confidence

Confidence score that image is of landmark

name

string

Landmark Name

Name of the identified landmark

GetAreaOfInterestParameterImage

Summary:

Description:

Properties:

Name

Type

Summary

GetImageSchemaResponse

Summary:

Description:

Properties:

Name

Type

Summary

schema

Schema

Summary:

Description:

Properties:

Name

Type

Summary

GetImageSchemaV2Response

Summary:

Description:

Properties:

Name

Type

Summary

schema

Schema

Summary:

Description:

Properties:

Name

Type

Summary

GetThumbnailParameterImage

Summary:

Description:

Properties:

Name

Type

Summary

GetThumbnailV2ParameterImage

Summary:

Description:

Properties:

Name

Type

Summary

ImageContentObject

Summary:

Description:

Properties:

Name

Type

Summary

url

string(url)

Image URL

Source of image reference by URL.

ListModelsResponse

Summary:

Description:

Properties:

Name

Type

Summary

models

array of (ModelResponse)

Models

Models.

ModelResponse

Summary:

Description:

Properties:

Name

Type

Summary

name

string

Model name

Name.

OCRJsonResponse

Summary: JSON Response

Description:

Properties:

Name

Type

Summary

language

string

Text Language

Detected language of the image text.

regions

array of (RegionsItem)

Regions Array

Text regions returned.

RegionsItem

Summary:

Description:

Properties:

Name

Type

Summary

OCRParameterImage

Summary:

Description:

Properties:

Name

Type

Summary

OCRTextParameterImage

Summary:

Description:

Properties:

Name

Type

Summary

OCRTextResponse

Summary:

Description:

Properties:

Name

Type

Summary

text

string

Detected Text

Text detected in the image analyzed

OCRTextV2ParameterImage

Summary:

Description:

Properties:

Name

Type

Summary

OCRV2ParameterImage

Summary:

Description:

Properties:

Name

Type

Summary

RecognizeDomainSpecificContentParameterImage

Summary:

Description:

Properties:

Name

Type

Summary

RecognizeDomainSpecificContentV2ParameterImage

Summary:

Description:

Properties:

Name

Type

Summary

TagImageParameterImage

Summary:

Description:

Properties:

Name

Type

Summary

TagImageV2ParameterImage

Summary:

Description:

Properties:

Name

Type

Summary

TagResponse

Summary:

Description:

Properties:

Name

Type

Summary

TagsItem

Summary:

Description:

Properties:

Name

Type

Summary

confidence

number(float)

Tag Confidence Score

Confidence score of the identified tag.

name

string

Tag Name

Name of the tag identified.