You are here: Start » AVL.NET » AVL.DeepLearning_ClassifyObject Method

AVL.DeepLearning_ClassifyObject Method

Performs whole image classification using a trained deep-learning model.

Namespace:AvlNet
Assembly:AVL.NET.dll

Syntax

public static void DeepLearning_ClassifyObject(
	ref AvlNet.DeepLearningConnectionState ioState,
	AvlNet.Image inImage,
	AvlNet.Region inRoi,
	string inDeepModel,
	out AvlNet.ClassConfidence[] outConfidences,
	out string outClassName,
	out int outClassIndex,
	out float outScore
)

Parameters

Name Type Range Default Description
ioStateAvlNet.DeepLearningConnectionState
inImageAvlNet.ImageInput image.
inRoiAvlNet.RegionArea of interest. Default value: atl::NIL, or null.
inDeepModelstringTrained model.
outConfidencesAvlNet.ClassConfidenceReturns confidences for all classes.
outClassNamestringReturns name of the class with highest confidence.
outClassIndexintReturns the index of the class with highest confidence.
outScorefloatReturns the value of the highest confidence.

Errors

Error type Description
DomainError Empty ROI region in DeepLearning_ClassifyObject
DomainError Invalid DeepModel in DeepLearning_ClassifyObject.
DomainError Empty image in DeepLearning_ClassifyObject.
IoError Unable to connect to Deep Learning Service.
IoError Connection with service lost.
IoError Missing results count.
IoError Missing confidences.

See also