/AWS1/CL_REK=>DETECTCUSTOMLABELS()
¶
About DetectCustomLabels¶
This operation applies only to Amazon Rekognition Custom Labels.
Detects custom labels in a supplied image by using an Amazon Rekognition Custom Labels model.
You specify which version of a model version to use by using the ProjectVersionArn
input
parameter.
You pass the input image as base64-encoded image bytes or as a reference to an image in an Amazon S3 bucket. If you use the AWS CLI to call Amazon Rekognition operations, passing image bytes is not supported. The image must be either a PNG or JPEG formatted file.
For each object that the model version detects on an image, the API returns a
(CustomLabel
) object in an array (CustomLabels
). Each
CustomLabel
object provides the label name (Name
), the level
of confidence that the image contains the object (Confidence
), and object
location information, if it exists, for the label on the image (Geometry
).
Note that for the DetectCustomLabelsLabels
operation, Polygons
are not returned in the Geometry
section of the response.
To filter labels that are returned, specify a value for MinConfidence
.
DetectCustomLabelsLabels
only returns labels with a confidence that's higher than
the specified value.
The value of MinConfidence
maps to the assumed threshold values
created during training. For more information, see Assumed threshold
in the Amazon Rekognition Custom Labels Developer Guide.
Amazon Rekognition Custom Labels metrics expresses an assumed threshold as a floating point value between 0-1. The range of
MinConfidence
normalizes the threshold value to a percentage value (0-100). Confidence
responses from DetectCustomLabels
are also returned as a percentage.
You can use MinConfidence
to change the precision and recall or your model.
For more information, see
Analyzing an image in the Amazon Rekognition Custom Labels Developer Guide.
If you don't specify a value for MinConfidence
, DetectCustomLabels
returns labels based on the assumed threshold of each label.
This is a stateless API operation. That is, the operation does not persist any data.
This operation requires permissions to perform the
rekognition:DetectCustomLabels
action.
For more information, see Analyzing an image in the Amazon Rekognition Custom Labels Developer Guide.
Method Signature¶
IMPORTING¶
Required arguments:¶
IV_PROJECTVERSIONARN
TYPE /AWS1/REKPROJECTVERSIONARN
/AWS1/REKPROJECTVERSIONARN
¶
The ARN of the model version that you want to use. Only models associated with Custom Labels projects accepted by the operation. If a provided ARN refers to a model version associated with a project for a different feature type, then an InvalidParameterException is returned.
IO_IMAGE
TYPE REF TO /AWS1/CL_REKIMAGE
/AWS1/CL_REKIMAGE
¶
Image
Optional arguments:¶
IV_MAXRESULTS
TYPE /AWS1/REKUINTEGER
/AWS1/REKUINTEGER
¶
Maximum number of results you want the service to return in the response. The service returns the specified number of highest confidence labels ranked from highest confidence to lowest.
IV_MINCONFIDENCE
TYPE /AWS1/RT_FLOAT_AS_STRING
/AWS1/RT_FLOAT_AS_STRING
¶
Specifies the minimum confidence level for the labels to return.
DetectCustomLabels
doesn't return any labels with a confidence value that's lower than this specified value. If you specify a value of 0,DetectCustomLabels
returns all labels, regardless of the assumed threshold applied to each label. If you don't specify a value forMinConfidence
,DetectCustomLabels
returns labels based on the assumed threshold of each label.