Package qupath.opencv.ops

Class ImageOps.ML

java.lang.Object

qupath.opencv.ops.ImageOps.ML

Enclosing class:

ImageOps

public static class ImageOps.ML
extends Object

Machine learning operations.

Constructor Summary

Constructors

Constructor	Description
ML()

Method Summary

Modifier and Type	Method	Description
static ImageOp	dnn(DnnModel model, int inputWidth, int inputHeight, Padding padding, String... outputNames)	Apply a DnnModel to pixels to generate a prediction.
static ImageOp	preprocessor(FeaturePreprocessor preprocessor)	Apply a FeaturePreprocessor to pixels, considering each channel as features.
static ImageOp	statModel(OpenCVClassifiers.OpenCVStatModel statModel, boolean requestProbabilities)	Apply a StatModel to pixels to generate a prediction.

Methods inherited from class java.lang.Object

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

Constructor Details

ML

public ML()

Method Details

statModel

public static ImageOp statModel(OpenCVClassifiers.OpenCVStatModel statModel,
 boolean requestProbabilities)

Apply a StatModel to pixels to generate a prediction.

Parameters:

statModel -

requestProbabilities -

Returns:

dnn

public static ImageOp dnn(DnnModel model,
 int inputWidth,
 int inputHeight,
 Padding padding,
 String... outputNames)

Apply a DnnModel to pixels to generate a prediction.

Parameters:

model -

inputWidth - requested input width

inputHeight - requested input height

padding - amount of padding provided

outputNames - names of model outputs. If empty, the first (and often only) output is used. If more than one output is specified, it is assumed that all are the same size and they be concatenated along the channels dimension.

Returns:

preprocessor

public static ImageOp preprocessor(FeaturePreprocessor preprocessor)

Apply a FeaturePreprocessor to pixels, considering each channel as features.

Parameters:

preprocessor -

Returns: