Human detection

Scenarios

• Security monitoring: Detect unauthorized individuals entering restricted or private areas. The system can automatically trigger alerts and notify security personnel.

• People counting: Count people in shopping malls, public venues, or large events to analyze crowding and support crowd management and evacuation planning.

Prerequisites

• You have created an AccessKey pair. For more information, see Create an AccessKey pair.

• You have activated OSS, created a bucket, and uploaded files to the bucket. For more information, see Upload files.

• You have activated IMM. For more information, see Activate IMM.

• You have created a project in the IMM console. For more information, see Create a project.

  Note

  • You can also call the CreateProject operation to create a project. For more information, see Create a project.

  • You can call the ListProjects operation to list all projects in a specified region.

Usage

Call the DetectImageBodies operation to detect humans in an image and retrieve each person's bounding box and confidence level.

{
  "ProjectName": "test-project",
  "SourceURI": "oss://test-bucket/test-object.jpg",
  "Sensitivity": 1
}

{
  "RequestId": "E9BDE106-1436-5077-9F6E-65D81DC6FEC8",
  "Bodies": [
    {
      "Confidence": 0.733,
      "Boundary": {
        "Left": 115,
        "Top": 97,
        "Height": 103,
        "Width": 63
      }
    },
    {
      "Confidence": 0.732,
      "Boundary": {
        "Left": 214,
        "Top": 121,
        "Height": 86,
        "Width": 63
      }
    },
    {
      "Confidence": 0.596,
      "Boundary": {
        "Left": 63,
        "Top": 102,
        "Height": 100,
        "Width": 82
      }
    },
    {
      "Confidence": 0.566,
      "Boundary": {
        "Left": 194,
        "Top": 123,
        "Height": 82,
        "Width": 40
      }
    }
  ]
}

# -*- coding: utf-8 -*-
import os

from alibabacloud_imm20200930.client import Client as imm20200930Client
from alibabacloud_tea_openapi import models as open_api_models
from alibabacloud_imm20200930 import models as imm_20200930_models
from alibabacloud_tea_util import models as util_models
from alibabacloud_tea_util.client import Client as UtilClient


def create_client(access_key_id, access_key_secret):
    """Initialize the IMM client with your AccessKey pair."""
    config = open_api_models.Config(
        access_key_id=access_key_id,
        access_key_secret=access_key_secret
    )
    # Specify the IMM endpoint.
    config.endpoint = 'imm.cn-beijing.aliyuncs.com'
    return imm20200930Client(config)


def main():
    # The AccessKey pair of an Alibaba Cloud account has permissions on all API operations.
    # To prevent security risks, we recommend that you call API operations or perform
    # routine O&M as a Resource Access Management (RAM) user.
    # We recommend that you do not include your AccessKey pair (AccessKey ID and AccessKey
    # secret) in your project code. Otherwise, the AccessKey pair may be leaked and the
    # security of all resources within your account may be compromised.
    # In this example, the AccessKey pair is read from environment variables to implement
    # identity verification for API access. For more information about how to configure
    # environment variables, see https://www.alibabacloud.com/help/document_detail/2361894.html.
    access_key_id = os.getenv("AccessKeyId")
    access_key_secret = os.getenv("AccessKeySecret")
    client = create_client(access_key_id, access_key_secret)

    request = imm_20200930_models.DetectImageBodiesRequest(
        project_name='test-project',
        source_uri='oss://test-bucket/test-object.jpg',
        sensitivity=1
    )
    runtime = util_models.RuntimeOptions()

    try:
        response = client.detect_image_bodies_with_options(request, runtime)

        # Process the detection results
        for body in response.body.bodies:
            print(f"Confidence: {body.confidence:.3f}, "
                  f"Position: ({body.boundary.left}, {body.boundary.top}), "
                  f"Size: {body.boundary.width}x{body.boundary.height}")
    except Exception as error:
        # Print the error message if necessary.
        UtilClient.assert_as_string(error.message)


if __name__ == '__main__':
    main()

Billing

Human detection incurs costs on both the OSS and IMM sides.

• OSS side: OSS Pricing.

  API	Billing item	Description
  GetObject	GET request	Billed by the number of successful requests.
  	Infrequent Access data retrieval	If the data uses the Infrequent Access storage class, retrieval fees apply based on the volume of retrieved data.
  	Real-time access of Archive objects	If the object uses Archive storage and the bucket has real-time access enabled, real-time access fees apply based on the retrieved data size.
  	Transfer acceleration	If transfer acceleration is enabled and you use an acceleration endpoint, transfer acceleration fees apply based on the data size.
  HeadObject	GET request	Billed by the number of successful requests.

• IMM side: IMM billing items.

  Important

  Starting from 11:00 UTC+8 on July 28, 2025, the IMM human detection service will merge into the image detection service, and the free tier will switch to a paid model. IMM billing adjustment announcement.

  API	Billing item	Description
  DetectImageBodies	ImageDetect	Billed by the number of successful requests.

References

Human detection