Image Tag

Last updated: 2025-09-09 20:02:21

Image Tag

Last updated: 2025-09-09 20:02:21

Feature Description
The image tagging feature uses the persistent processing API of Cloud Infinite to implement image tagging recognition for existing COS data, returning high-confidence topic tags in the images.
Note:
This API is a GET request and uses a synchronous request method. It requires carrying a signature. For specific signature settings, please see Request Signature.
Authorization Description
When using with a sub-account, you need to add ci:CreateDetectLabelJob in the action of the authorization policy. For details on CI access permission, please refer to Cloud Infinite action.
Service Activation
Using this feature requires enabling Cloud Infinite in advance and binding a bucket.
Using this feature requires enabling the AI Content Recognition Service in advance through the console or API.
Use Limits
When using this API, please confirm the relevant restrictions first. For details, see Usage Limits.
Fee Description
This API is a paid service. Generated costs will be charged by Cloud Infinite. For detailed billing instructions, see Content Recognition.
﻿
Request
Original images are stored in COS.
GET /<ObjectKey>?ci-process=detect-label&scenes=<scenes> HTTP/1.1
Host: <BucketName-APPID>.cos.<Region>.myqcloud.com
Date: <GMT Date>
Authorization: <Auth String>
Original images come from other links.
GET /?ci-process=detect-label&scenes=<scenes>&detect-url=<detect-url> HTTP/1.1
Host: <BucketName-APPID>.cos.<Region>.myqcloud.com
Date: <GMT Date>
Authorization: <Auth String>
Note:
Authorization: Auth String. For details, see Request Signature document.
Request Content
Parameter Name
Description
Type
Required or Not
ObjectKey
object filename, such as picture.jpg
string
No
ci-process
fixed value detect-label
string
Yes
scenes
Supported recognition scenarios for this call, with the following options:
web Image optimization for networks
camera image optimization for mobile camera shooting
{mobile phone album} {cloud storage product} {product optimization}
Optimized for news, information, broadcasting and television industries
if this parameter is not provided, default to camera
Support multi-scene (scenes) detection together, separated by ,. For example, use scenes = web, camera to apply two models to detect one image simultaneously, outputting two sets of recognition results.
String
No
detect-url
You can process any publicly accessible image link by filling in detect-url. If detect-url is not filled in, the backend will process the ObjectKey by default. If detect-url is filled in, the backend will process the detect-url link, and there is no need to fill in ObjectKey. detect-url example: http://www.example.com/abc.jpg, which needs to be url-encoded and processed into http%25253A%25252F%25252Fwww.example.com%25252Fabc.jpg.
String
No
Response
The specific data content of the response body is as follows:
Parameter Name
Description
Type
RecognitionResult
Image tag recognition results
Container
Content of the RecognitionResult node
Parameter Name
Description
Type
AlbumLabels
Album edition tag result array
If the album scenario is not selected, this item does not exist. This field may be empty, indicating that no valid value can be obtained.
Container
CameraLabels
Camera edition tag result array
If the camera scenario is not selected, this item does not exist. This field may be empty, indicating that no valid value can be obtained.
Container
NewsLabels
News edition tag result array.
If the news scenario is not selected, this item does not exist. This field may be empty, indicating that no valid value can be obtained.
Container
WebLabels
Web edition tag result array
If the web scenario is not selected, this item does not exist. This field may be empty, indicating that no valid value can be obtained.
Container
WebLabels, CameraLabels, AlbumLabels, NewsLabels
Parameter Name
Description
Type
Labels
Tag information
Container
Content of the Labels node
Parameter Name
Description
Type
Confidence
The confidence score of this tag. The higher the score, the higher the accuracy of the tag.
Int
FirstCategory
Primary category of the tag
String
Name
Recognized image tags
String
SecondCategory
Secondary category of a tag
String
Example
Request
GET /picture.jpg?ci-process=detect-label&scenes=web,camera,album,news HTTP/1.1
Host: examplebucket-1250000000.cos.ap-chengdu.myqcloud.com
Date: Tue, 03 Apr 2019 09:06:15 GMT
Authorization: <Auth String>
Response
HTTP/1.1 200 OK
Content-Type: application/xml
Content-Length: 645
Date: Tue, 03 Apr 2018 09:06:16 GMT
Status: 200 OK
x-cos-request-id: NWFjMzQ0MDZfOTBmYTUwXzZkZV8z****
﻿
<RecognitionResult>
    <AlbumLabels>
        <Labels>
            <Confidence>72</Confidence>
            <FirstCategory>Animals and plants</FirstCategory>
            <Name>Deer</Name>
            <SecondCategory>Mammals</SecondCategory>
        </Labels>
    </AlbumLabels>
    <CameraLabels>
        <Labels>
            <Confidence>26</Confidence>
            <FirstCategory>Animals and plants</FirstCategory>
            <Name>Antelope</Name>
            <SecondCategory>Mammals</SecondCategory>
        </Labels>
        <Labels>
            <Confidence>21</Confidence>
            <FirstCategory>Animals and plants</FirstCategory>
            <Name>Rabbit</Name>
            <SecondCategory>Mammals</SecondCategory>
        </Labels>
        <Labels>
            <Confidence>18</Confidence>
            <FirstCategory>Animals and plants</FirstCategory>
            <Name>Deer</Name>
            <SecondCategory>Mammals</SecondCategory>
        </Labels>
        <Labels>
            <Confidence>10</Confidence>
            <FirstCategory>Animals and plants</FirstCategory>
            <Name>Lion</Name>
            <SecondCategory>Mammals</SecondCategory>
        </Labels>
        <Labels>
            <Confidence>5</Confidence>
            <FirstCategory>Animals and plants</FirstCategory>
            <Name>Leopard</Name>
            <SecondCategory>Mammals</SecondCategory>
        </Labels>
    </CameraLabels>
    <NewsLabels>
        <Labels>
            <Confidence>60</Confidence>
            <FirstCategory>Plants and other organisms</FirstCategory>
            <Name>Plant</Name>
            <SecondCategory>Plant category</SecondCategory>
        </Labels>
        <Labels>
            <Confidence>38</Confidence>
            <FirstCategory>Plant</FirstCategory>
            <Name>Tree</Name>
            <SecondCategory>Tree</SecondCategory>
        </Labels>
    </NewsLabels>
    <WebLabels>
        <Labels>
            <Confidence>97</Confidence>
            <FirstCategory>Animals and plants</FirstCategory>
            <Name>Wildlife</Name>
            <SecondCategory>Other</SecondCategory>
        </Labels>
        <Labels>
            <Confidence>82</Confidence>
            <FirstCategory>Animals and plants</FirstCategory>
            <Name>Mammals</Name>
            <SecondCategory>Mammals</SecondCategory>
        </Labels>
        <Labels>
            <Confidence>78</Confidence>
            <FirstCategory>Animals and plants</FirstCategory>
            <Name>Animal</Name>
            <SecondCategory>Other</SecondCategory>
        </Labels>
        <Labels>
            <Confidence>71</Confidence>
            <FirstCategory>Animals and plants</FirstCategory>
            <Name>Deer</Name>
            <SecondCategory>Mammals</SecondCategory>
        </Labels>
        <Labels>
            <Confidence>41</Confidence>
            <FirstCategory>Animals and plants</FirstCategory>
            <Name>White-tailed Deer</Name>
            <SecondCategory>Mammals</SecondCategory>
        </Labels>
    </WebLabels>
</RecognitionResult>
﻿

Was this page helpful?

You can also Contact Sales or Submit a Ticket for help.

Yes

Feedback

Parameter Name	Description	Type	Required or Not
ObjectKey	object filename, such as picture.jpg	string	No
ci-process	fixed value detect-label	string	Yes
scenes	Supported recognition scenarios for this call, with the following options: web Image optimization for networks camera image optimization for mobile camera shooting {mobile phone album} {cloud storage product} {product optimization} Optimized for news, information, broadcasting and television industries if this parameter is not provided, default to camera Support multi-scene (scenes) detection together, separated by `,`. For example, use scenes = web, camera to apply two models to detect one image simultaneously, outputting two sets of recognition results.	String	No
detect-url	You can process any publicly accessible image link by filling in detect-url. If detect-url is not filled in, the backend will process the ObjectKey by default. If detect-url is filled in, the backend will process the detect-url link, and there is no need to fill in ObjectKey. detect-url example: `http://www.example.com/abc.jpg`, which needs to be url-encoded and processed into `http%25253A%25252F%25252Fwww.example.com%25252Fabc.jpg`.	String	No

tencent cloud

Feature Description

Authorization Description

Service Activation

Use Limits

Fee Description

Request

Request Content

Response

Example

Request

Response