All Products
Search
Document Center

Object Storage Service:Image intelligence overview

Last Updated:Jun 18, 2026

Image intelligence uses AI algorithms to analyze image content, including facial detection, tag detection, and quality assessment. You can call the x-oss-process and x-oss-async-process operations to analyze images stored in Object Storage Service (OSS) from any connected device.

Parameters

Each image intelligence feature is controlled by a single parameter. You can encapsulate the parameter in a style to process images in bulk. For more information, see Styles.

The following table describes the image intelligence parameters:

Image intelligence

Parameter

Description

Facial detection

image/faces

Detects the position of faces in the image and analyzes facial attributes.

Body detection

image/bodies

Detects the position of bodies in the image.

Vehicle detection

image/cars

Detects and analyzes information about vehicles and license plates in the image.

QR code recognition

image/codes

Identifies QR codes in the image.

Image tag detection

image/labels

Identifies tags associated with scenes, objects, and events in the image.

Image quality assessment

image/score

Assesses the aesthetic quality of the image based on metrics such as color and saturation.

Blind watermarking

image/blindwatermark

Adds text blind watermarks to an image.

image/deblindwatermark

Decodes text blind watermarks from an image.

Usage

Process images by calling the x-oss-process operation. For more information, see Synchronous processing.

To decode blind watermarks, call the x-oss-async-process operation. For more information, see Asynchronous processing.

Billing

Image intelligence fees are included in Intelligent Media Management (IMM) bills. For more information, see billable items.