Pangu-CV-ImageClassification-2.1.0

Function

Quantitatively analyzes an image based on different image features and classifies the image into several categories. It is applicable to tasks such as animal and plant classification, vehicle type classification, license plate classification, scrap grading, and component classification.

Service API calling method:

Image inference is supported.
Image inference services can be deployed as real-time services or edge services.

Authorization Information

An account has required permissions to call all APIs by default. To call this API as an IAM user, the IAM user must be granted the required permissions. For details, see Permissions and Supported Actions.

URI

POST /v1/{project_id}/infer-api/proxy/service/{deployment_id}/

For details about how to obtain the URI, see Request URI.

**Table 1** Path parameters of the inference API
Parameter	Mandatory	Type	Description
project_id	Yes	String	Definition: Project ID. For details about how to obtain a project ID, see Obtaining the Project ID. Constraints: N/A Value range: N/A Default value: N/A
deployment_id	Yes	String	Definition: Model deployment ID. For details about how to obtain the deployment ID, see Obtaining the Model Deployment ID. Constraints: N/A Value range: N/A Default value: N/A

Request Parameters

Table 2 lists the request header parameters for token-based authentication.

**Table 2** Request header parameters (token-based authentication)
Parameter	Mandatory	Type	Description
X-Auth-Token	Yes	String	Definition: User token. Used to obtain the permission required to call APIs. The token is the value of X-Subject-Token in the response header in Figure 4. Constraints: N/A Value range: N/A Default value: N/A
Content-Type	Yes	String	Definition: MIME type of the body in the request. Constraints: N/A Value range: N/A Default value: application/json

Table 3 lists the request header parameters for API key authentication.

**Table 3** Request header parameters (API key authentication)
Parameter	Mandatory	Type	Description
X-Apig-AppCode	Yes	String	Definition: API key. Used to obtain the permission required to call APIs. The API key is the value of X-Apig-AppCode in the response header in API key authentication. Constraints: N/A Value range: N/A Default value: N/A
Content-Type	Yes	String	Definition: MIME type of the body in the request. Constraints: N/A Value range: N/A Default value: application/json

**Table 4** Request body parameters
Parameter	Mandatory	Type	Description
images	Yes	String/List[String]	Definition: Base64-encoded image. Constraints: It is recommended that the maximum size of a request body is 4 MB. You are advised to use images in JPG, PNG, JPEG, or BMP format. By default, only the RGB three-channel images are supported. If a single image is contained in the request, the parameter type is String, which is the Base64 code of the image. If multiple images are contained in the request, the parameter type is List[String], which stores the Base64 codes of all images in the list. A maximum of 24 images can be contained in a request. Value range: N/A Default value: N/A
mode	No	String	Definition: The value can be "single" or "multiple", indicating single-label classification or multi-label classification, respectively. The default value is the mode corresponding to the trained model. Constraints: N/A Value range: single: single-label classification multiple: multi-label classification Default value: N/A
threshold	No	dict	Definition: Prediction score threshold of each label in multi-label classification. Predictions whose scores are less than the threshold will be filtered out. Constraints: This parameter is available only in multi-label classification mode. Value range: N/A Default value: N/A
top	No	int	Definition: Predictions whose scores rank top N in single-label classification. Constraints: This parameter is available only in single-label classification mode. Value range: N/A Default value: N/A

Response Parameters

Status code: 200

If the response is successful, the returned structure is a dict, which consists of the predictions of multiple input images in the current request. The images are distinguished by numbers (keys).

**Table 5** Parameters in the body of a successful response to a single-or multi-label classification request
Parameter	Type	Description
Key	String	Definition: Sequence numbers of the input images. The value starts from 0 and cannot exceed 23. Constraints: N/A Value range: 0-23 Default value: N/A
Value	List[Dict]	Definition: Prediction result corresponding to the image with the current number. Constraints: N/A Value range: N/A Default value: N/A
dataset_id	String	Definition: Training dataset ID Constraints: N/A Value range: N/A Default value: N/A

The prediction result parameter type of each image is List[Dict], indicating that one or more classes have been predicted. For details about the parameter content of each dict, see Table 6.

**Table 6** Single-class prediction result parameters of a single image
Parameter	Type	Description
label	String	Definition: Predicted label, which is the same as the label defined in the training data Constraints: N/A Value range: N/A Default value: N/A
score	String	Definition: Confidence of the prediction. The prediction score of each label is provided. The score ranges from 0 to 1. Constraints: N/A Value range: 0-1 Default value: N/A

Status code: 400

**Table 7** Parameters in the response body for a failed request
Parameter	Type	Description
error_code	String	Error code
error_msg	String	Error message

Example Request

Example of a single-image classification request:

{
  "images": "/9j/4Vr2RXhpZgAASUkqAAgAAA.....",
}

Example of a multi-image classification request (maximum number of images that can be contained in a request: 24):

{
  "images": ["/9j/4Vr2RXhpZgAASUkqAAgAAA.....", "/9j/4RIrRXhpZgAATU......"]
}

Example of a single-label classification request with advanced parameters:

{
  "images": ["/9j/4Vr2RXhpZgAASUkqAAgAAA.....", "/9j/4RIrRXhpZgAATU......"],
  "top": 3
}

Example of a multi-label classification request with advanced parameters:

{
  "images": ["/9j/4Vr2RXhpZgAASUkqAAgAAA.....", "/9j/4RIrRXhpZgAATU......"],
  "threshold": 
  {
    "bird": 0.33,
    "blackbird": 0.44
  }
}

Example Response

A dictionary is returned in the response. The Key is the numbers of the input images of the current request. The input images are numbered from 0 in sequence. The Value is a list, which contains the prediction results of the images. Each image may have multiple prediction results (for example, in multi-label classification mode).

{
    "0": [
        {
            "label": "bird",
            "score": "0.95511043"
        },
        {
            "label": "blackbird",
            "score": "0.75241840"
        },
    ],
    "1": [
        {
            "label": "bird",
            "score": "0.36211243"
        }
    ],
    "dataset_id": "1341002014632579072"
}