Package org.anchoranalysis.plugin.onnx.bean.object.segment.decode.instance.text

Class DecodeEAST

Object

AnchorBean<DecodeInstanceSegmentation<ai.onnxruntime.OnnxTensor>>

DecodeInstanceSegmentation<ai.onnxruntime.OnnxTensor>

DecodeEAST

public class DecodeEAST
extends DecodeInstanceSegmentation<ai.onnxruntime.OnnxTensor>

Extracts text from a RGB image by using the EAST deep neural network model and the ONNX Runtime.

Each object-mask represented rotated-bounding box and is associated with a confidence score.

Particular thanks to Adrian Rosebrock whose tutorial was useful in applying this model

Author:

Owen Feehan

Constructor Summary

Constructors

Constructor	Description
DecodeEAST()

Method Summary

Modifier and Type	Method	Description
List<LabelledWithConfidence<MultiScaleObject>>	decode(List<ai.onnxruntime.OnnxTensor> inferenceOutput, ImageInferenceContext context)	Decodes the output tensors from inference into ObjectMasks with confidence and labels.
List<String>	expectedOutputs()	Ordered names of the tensors we are interested in processing, as outputted from inference.
double	getMinConfidence()	Proposed bounding boxes below this confidence interval are removed from consideration.
void	setMinConfidence(double minConfidence)	Proposed bounding boxes below this confidence interval are removed from consideration.

Methods inherited from class org.anchoranalysis.bean.AnchorBean

checkMisconfigured, describeBean, describeChildren, duplicateBean, fields, findFieldsOfClass, getBeanName, getLocalPath, localise, toString

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, wait, wait, wait

Constructor Details

DecodeEAST

public DecodeEAST()

Method Details

decode

public List<LabelledWithConfidence<MultiScaleObject>> decode(List<ai.onnxruntime.OnnxTensor> inferenceOutput,
 ImageInferenceContext context)
                                                      throws OperationFailedException

Description copied from class: DecodeInstanceSegmentation

Decodes the output tensors from inference into ObjectMasks with confidence and labels.

The created ObjectMasks should match unscaledDimensions in size.

Specified by:

decode in class DecodeInstanceSegmentation<ai.onnxruntime.OnnxTensor>

Parameters:

inferenceOutput - the tensors that are the result of the inference.

context - the context in which the inference is occurring.

Returns:

a newly created list of objects, with associated confidence, and labels, that matches unscaledDimensions in size.

Throws:

OperationFailedException - if it cannot be decoded successfully.

expectedOutputs

public List<String> expectedOutputs()

Description copied from class: DecodeInstanceSegmentation

Ordered names of the tensors we are interested in processing, as outputted from inference.

Specified by:

expectedOutputs in class DecodeInstanceSegmentation<ai.onnxruntime.OnnxTensor>

Returns:

the list of names, as above.

getMinConfidence

public double getMinConfidence()

Proposed bounding boxes below this confidence interval are removed from consideration.

setMinConfidence

public void setMinConfidence(double minConfidence)

Proposed bounding boxes below this confidence interval are removed from consideration.