CmdID / aws / aws rekognition

aws rekognition detect-labels

Detects instances of real-world entities within an image (JPEG or PNG) provided as input. AWS CLI reference

1 example 5 options Its page in the AWS CLI reference

Examples

from aws-cli, each command read by CmdID

To detect a label in an image

The following detect-labels example detects scenes and objects in an image stored in an Amazon S3 bucket.

aws rekognition detect-labels --image '{"S3Object":{"Bucket":"amzn-s3-demo-bucket","Name":"image"}}'Identify

--imageThe input image as base64-encoded bytes or an S3 object.

Its output, shortened

{
    "Labels": [
        {
            "Instances": [],
            "Confidence": 99.15271759033203,
            "Parents": [
                {
                    "Name": "Vehicle"
                },
                {
                    "Name": "Transportation"
                }
            ],
            "Name": "Automobile"
        },
        {
            "Instances": [],
            "Confidence": 99.15271759033203,
            "Parents": [
                {
                    "Name": "Transportation"
                }
            ],
            "Name": "Vehicle"
        },
        {
            "Instances": [],
            "Confidence": 99.15271759033203,
            "Parents": [],
            "Name": "Transportation"
… 523 more lines

For more information, see Detecting Labels in an Image in the Amazon Rekognition Developer Guide.

Options

5 in the AWS CLI reference

--featuresA list of the types of analysis to perform.

--imageThe input image as base64-encoded bytes or an S3 object.

--max-labelsMaximum number of labels you want the service to return in the response.

--min-confidenceSpecifies the minimum confidence level for the labels to return.

--settingsA list of the filters to be applied to returned detected labels and image properties.