org.openimaj.tools.ocr

Class KNNCharacterClassifier.ImageTrainer

java.lang.Object

org.openimaj.tools.ocr.KNNCharacterClassifier.ImageTrainer

All Implemented Interfaces:

FontSimulator.FontSimListener<FImage>

Enclosing class:

KNNCharacterClassifier

protected class KNNCharacterClassifier.ImageTrainer
extends Object
implements FontSimulator.FontSimListener<FImage>

This class is the FontSimulator.FontSimListener implementation that receives each character as an image in some randomised font. When it receives an image, it creates the feature vector by calling KNNCharacterClassifier.getImageVector(FImage) and stores that into a double array that can be used as input to the nearest neighbour classifier for training. That array can be retrieved using getVector().

Author:

David Dupplaw (dpd@ecs.soton.ac.uk)

Created:

19 Aug 2011

Constructor Summary

Constructors

Constructor and Description
ImageTrainer(int n) Constructor that takes the size of the output vector (that is the total number of training instances).

Method Summary

Modifier and Type	Method and Description
double[][]	getVector() Retrieve the set of training data as a double array.
void	imageCreated(FImage img) Implementation of the FontSimulator.FontSimListener interface.

Methods inherited from class java.lang.Object

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

Constructor Detail

ImageTrainer

public ImageTrainer(int n)

Constructor that takes the size of the output vector (that is the total number of training instances).

Parameters:

n - The number of training instances

Method Detail

imageCreated

public void imageCreated(FImage img)

Implementation of the FontSimulator.FontSimListener interface.

Specified by:

imageCreated in interface FontSimulator.FontSimListener<FImage>

Parameters:

img - The character image.

getVector

public double[][] getVector()

Retrieve the set of training data as a double array.

Returns:

the training data