Package com.itextpdf.pdfocr.onnx.detection

Class OnnxDetectionPredictorProperties

java.lang.Object

com.itextpdf.pdfocr.onnx.AbstractOnnxPredictorProperties

com.itextpdf.pdfocr.onnx.detection.OnnxDetectionPredictorProperties

       public class OnnxDetectionPredictorProperties extends AbstractOnnxPredictorProperties
      

Properties for configuring text detection ONNX models.

It contains a path to the model, model input properties and a model output post-processor.

Field Summary

Fields inherited from class com.itextpdf.pdfocr.onnx.AbstractOnnxPredictorProperties

DEFAULT_ORT_SESSION_CREATOR, inputProperties, modelPath, ortSessionOptionsCreator

Constructor Summary

Constructors

Constructor	Description
OnnxDetectionPredictorProperties(String modelPath, OnnxInputProperties inputProperties, IDetectionPostProcessor postProcessor)	Creates new text detection predictor properties.
OnnxDetectionPredictorProperties(String modelPath, OnnxInputProperties inputProperties, IDetectionPostProcessor postProcessor, IOrtSessionOptionsCreator ortSessionOptionsCreator)	Creates new text detection predictor properties.

Method Summary

Modifier and Type	Method	Description
static OnnxDetectionPredictorProperties	dbNet(String modelPath)	Creates a new text detection properties object for existing pre-trained DBNet models, stored on disk.
static OnnxDetectionPredictorProperties	dbNet(String modelPath, IOrtSessionOptionsCreator ortSessionOptionsCreator)	Creates a new text detection properties object for existing pre-trained DBNet models, stored on disk.
static OnnxDetectionPredictorProperties	easyOcr(String modelPath)	Creates a new text detection properties object for an existing pre-trained EasyOCR CRAFT model, stored on disk.
static OnnxDetectionPredictorProperties	easyOcr(String modelPath, IOrtSessionOptionsCreator ortSessionOptionsCreator)	Creates a new text detection properties object for an existing pre-trained EasyOCR CRAFT model, stored on disk.
boolean	equals(Object o)
static OnnxDetectionPredictorProperties	fast(String modelPath)	Creates a new text detection properties object for existing pre-trained FAST models, stored on disk.
static OnnxDetectionPredictorProperties	fast(String modelPath, IOrtSessionOptionsCreator ortSessionOptionsCreator)	Creates a new text detection properties object for existing pre-trained FAST models, stored on disk.
IDetectionPostProcessor	getPostProcessor()	Returns the ONNX model output post-processor.
int	hashCode()
static OnnxDetectionPredictorProperties	linkNet(String modelPath)	Creates a new text detection properties object for existing pre-trained LinkNet models, stored on disk.
static OnnxDetectionPredictorProperties	linkNet(String modelPath, IOrtSessionOptionsCreator ortSessionOptionsCreator)	Creates a new text detection properties object for existing pre-trained LinkNet models, stored on disk.
static OnnxDetectionPredictorProperties	paddleOcr(String modelDirPath)	Creates a new text detection properties object for existing pre-trained PaddleOCR models, stored on disk.
static OnnxDetectionPredictorProperties	paddleOcr(String modelDirPath, IOrtSessionOptionsCreator ortSessionOptionsCreator)	Creates a new text detection properties object for existing pre-trained PaddleOCR models, stored on disk.
static OnnxDetectionPredictorProperties	paddleOcr(String modelPath, String configPath)	Creates a new text detection properties object for existing pre-trained PaddleOCR models, stored on disk.
static OnnxDetectionPredictorProperties	paddleOcr(String modelPath, String configPath, IOrtSessionOptionsCreator ortSessionOptionsCreator)	Creates a new text detection properties object for existing pre-trained PaddleOCR models, stored on disk.
String	toString()

Methods inherited from class com.itextpdf.pdfocr.onnx.AbstractOnnxPredictorProperties

getInputProperties, getModelPath, getOrtSessionOptionsCreator

Methods inherited from class java.lang.Object

clone, finalize, getClass, notify, notifyAll, wait, wait, wait

Constructor Details

OnnxDetectionPredictorProperties

             public OnnxDetectionPredictorProperties
             (String modelPath, OnnxInputProperties inputProperties, IDetectionPostProcessor postProcessor)
            

Creates new text detection predictor properties.

Parameters:

modelPath - path to the ONNX model to load

inputProperties - ONNX model input properties

postProcessor - ONNX model output post-processor

OnnxDetectionPredictorProperties

             public OnnxDetectionPredictorProperties
             (String modelPath, OnnxInputProperties inputProperties, IDetectionPostProcessor postProcessor, IOrtSessionOptionsCreator ortSessionOptionsCreator)
            

Creates new text detection predictor properties.

Parameters:

modelPath - path to the ONNX model to load

inputProperties - ONNX model input properties

postProcessor - ONNX model output post-processor

ortSessionOptionsCreator - ONNX runtime session options creator

Method Details

dbNet

             public static OnnxDetectionPredictorProperties dbNet
             (String modelPath)
            

Creates a new text detection properties object for existing pre-trained DBNet models, stored on disk.

This can be used to load the following models from OnnxTR:

• db_resnet50
• db_resnet50 (8-bit quantized)
• db_resnet34
• db_resnet34 (8-bit quantized)
• db_mobilenet_v3_large
• db_mobilenet_v3_large (8-bit quantized)

These models output boxes of words.

Parameters:

modelPath - path to the pre-trained model

Returns:

a new text detection properties object for a DBNet model

dbNet

             public static OnnxDetectionPredictorProperties dbNet
             (String modelPath, IOrtSessionOptionsCreator ortSessionOptionsCreator)
            

Creates a new text detection properties object for existing pre-trained DBNet models, stored on disk.

This can be used to load the following models from OnnxTR:

• db_resnet50
• db_resnet50 (8-bit quantized)
• db_resnet34
• db_resnet34 (8-bit quantized)
• db_mobilenet_v3_large
• db_mobilenet_v3_large (8-bit quantized)

These models output boxes of words.

Parameters:

modelPath - path to the pre-trained model

ortSessionOptionsCreator - the ONNX runtime session options creator

Returns:

a new text detection properties object for a DBNet model

fast

             public static OnnxDetectionPredictorProperties fast
             (String modelPath)
            

Creates a new text detection properties object for existing pre-trained FAST models, stored on disk. This is the default text detection model in OnnxTR.

This can be used to load the following models from OnnxTR:

• fast_base
• fast_small
• fast_tiny

These models output boxes of words.

Parameters:

modelPath - path to the pre-trained model

Returns:

a new text detection properties object for a FAST model

fast

             public static OnnxDetectionPredictorProperties fast
             (String modelPath, IOrtSessionOptionsCreator ortSessionOptionsCreator)
            

Creates a new text detection properties object for existing pre-trained FAST models, stored on disk. This is the default text detection model in OnnxTR.

This can be used to load the following models from OnnxTR:

• fast_base
• fast_small
• fast_tiny

These models output boxes of words.

Parameters:

modelPath - path to the pre-trained model

ortSessionOptionsCreator - the ONNX runtime session options creator

Returns:

a new text detection properties object for a FAST model

linkNet

             public static OnnxDetectionPredictorProperties linkNet
             (String modelPath)
            

Creates a new text detection properties object for existing pre-trained LinkNet models, stored on disk.

This can be used to load the following models from OnnxTR:

• linknet_resnet50
• linknet_resnet50 (8-bit quantized)
• linknet_resnet34
• linknet_resnet34 (8-bit quantized)
• linknet_resnet18
• linknet_resnet18 (8-bit quantized)

These models output boxes of words.

Parameters:

modelPath - path to the pre-trained model

Returns:

a new text detection properties object for a LinkNet model

linkNet

             public static OnnxDetectionPredictorProperties linkNet
             (String modelPath, IOrtSessionOptionsCreator ortSessionOptionsCreator)
            

Creates a new text detection properties object for existing pre-trained LinkNet models, stored on disk.

This can be used to load the following models from OnnxTR:

• linknet_resnet50
• linknet_resnet50 (8-bit quantized)
• linknet_resnet34
• linknet_resnet34 (8-bit quantized)
• linknet_resnet18
• linknet_resnet18 (8-bit quantized)

These models output boxes of words.

Parameters:

modelPath - path to the pre-trained model

ortSessionOptionsCreator - the ONNX runtime session options creator

Returns:

a new text detection properties object for a LinkNet model

paddleOcr

             public static OnnxDetectionPredictorProperties paddleOcr
             (String modelDirPath) throws IOException
            

Creates a new text detection properties object for existing pre-trained PaddleOCR models, stored on disk.

Only models in the ONNX format are supported. Since, by default, PaddleOCR does not provide models in the ONNX format, you might need to do a model conversion yourself. Check out this page for information on how to do that.

This method expects the directory to contain two files:

• inference.onnx - the inference model in the ONNX format
• inference.yml - the configuration file for the model in YAML

This method can be used to load the following PaddleOCR models:

• PP-OCRv5_server_det
• PP-OCRv5_mobile_det
• PP-OCRv4_server_det
• PP-OCRv4_mobile_det

These models output boxes of text lines. Make sure you choose a recognition model that can handle spaces.

Parameters:

modelDirPath - path to the directory with the model and its configuration file

Returns:

a new text detection properties object for a PaddleOCR model

Throws:

IOException - if any I/O error occurs while loading configuration file

paddleOcr

             public static OnnxDetectionPredictorProperties paddleOcr
             (String modelDirPath, IOrtSessionOptionsCreator ortSessionOptionsCreator) throws IOException
            

Creates a new text detection properties object for existing pre-trained PaddleOCR models, stored on disk.

Only models in the ONNX format are supported. Since, by default, PaddleOCR does not provide models in the ONNX format, you might need to do a model conversion yourself. Check out this page for information on how to do that.

This method expects the directory to contain two files:

• inference.onnx - the inference model in the ONNX format
• inference.yml - the configuration file for the model in YAML

This method can be used to load the following PaddleOCR models:

• PP-OCRv5_server_det
• PP-OCRv5_mobile_det
• PP-OCRv4_server_det
• PP-OCRv4_mobile_det

These models output boxes of text lines. Make sure you choose a recognition model that can handle spaces.

Parameters:

modelDirPath - path to the directory with the model and its configuration file

ortSessionOptionsCreator - the ONNX runtime session options creator

Returns:

a new text detection properties object for a PaddleOCR model

Throws:

IOException - if any I/O error occurs while loading configuration file

paddleOcr

             public static OnnxDetectionPredictorProperties paddleOcr
             (String modelPath, String configPath) throws IOException
            

Creates a new text detection properties object for existing pre-trained PaddleOCR models, stored on disk.

Only models in the ONNX format are supported. Since, by default, PaddleOCR does not provide models in the ONNX format, you might need to do a model conversion yourself. Check out this page for information on how to do that.

This method can be used to load the following PaddleOCR models:

• PP-OCRv5_server_det
• PP-OCRv5_mobile_det
• PP-OCRv4_server_det
• PP-OCRv4_mobile_det

These models output boxes of text lines. Make sure you choose a recognition model that can handle spaces.

Parameters:

modelPath - path to the pre-trained model in the ONNX format

configPath - path to the configuration file for the model

Returns:

a new text detection properties object for a PaddleOCR model

Throws:

IOException - if any I/O error occurs while loading configuration file

paddleOcr

             public static OnnxDetectionPredictorProperties paddleOcr
             (String modelPath, String configPath, IOrtSessionOptionsCreator ortSessionOptionsCreator) throws IOException
            

Creates a new text detection properties object for existing pre-trained PaddleOCR models, stored on disk.

Only models in the ONNX format are supported. Since, by default, PaddleOCR does not provide models in the ONNX format, you might need to do a model conversion yourself. Check out this page for information on how to do that.

This method can be used to load the following PaddleOCR models:

• PP-OCRv5_server_det
• PP-OCRv5_mobile_det
• PP-OCRv4_server_det
• PP-OCRv4_mobile_det

These models output boxes of text lines. Make sure you choose a recognition model that can handle spaces.

Parameters:

modelPath - path to the pre-trained model in the ONNX format

configPath - path to the configuration file for the model

ortSessionOptionsCreator - the ONNX runtime session options creator

Returns:

a new text detection properties object for a PaddleOCR model

Throws:

IOException - if any I/O error occurs while loading configuration file

easyOcr

             public static OnnxDetectionPredictorProperties easyOcr
             (String modelPath)
            

Creates a new text detection properties object for an existing pre-trained EasyOCR CRAFT model, stored on disk.

Only models in the ONNX format are supported. Since, by default, EasyOCR does not provide models in the ONNX format, you might need to do a model conversion yourself.

This can be used to load the following models from EasyOCR:

• CRAFT

These models output boxes of text lines. Make sure you choose a recognition model that can handle spaces.

Parameters:

modelPath - path to the pre-trained model

Returns:

a new text detection properties object for an EasyOCR CRAFT model

easyOcr

             public static OnnxDetectionPredictorProperties easyOcr
             (String modelPath, IOrtSessionOptionsCreator ortSessionOptionsCreator)
            

Creates a new text detection properties object for an existing pre-trained EasyOCR CRAFT model, stored on disk.

Only models in the ONNX format are supported. Since, by default, EasyOCR does not provide models in the ONNX format, you might need to do a model conversion yourself.

This can be used to load the following models from EasyOCR:

• CRAFT

These models output boxes of text lines. Make sure you choose a recognition model that can handle spaces.

Parameters:

modelPath - path to the pre-trained model

ortSessionOptionsCreator - the ONNX runtime session options creator

Returns:

a new text detection properties object for an EasyOCR CRAFT model

getPostProcessor

             public IDetectionPostProcessor getPostProcessor()
            

Returns the ONNX model output post-processor.

Returns:

the ONNX model output post-processor

equals

             public boolean equals
             (Object o)
            

Overrides:

equals in class Object

hashCode

             public int hashCode()
            

Overrides:

hashCode in class Object

toString

             public String toString()
            

Overrides:

toString in class Object