net.sourceforge.javaocr.ocrPlugins.mseOCR
Class TrainingImageLoader

java.lang.Object
  net.sourceforge.javaocr.scanner.DocumentScannerListenerAdaptor
      net.sourceforge.javaocr.ocrPlugins.mseOCR.TrainingImageLoader

All Implemented Interfaces:

DocumentScannerListener

public class TrainingImageLoader
extends DocumentScannerListenerAdaptor

Utility class to load an image file, break it into individual characters, and create and store multiple TrainingImage objects from the characters. The source image must contain a range of characters specified in the character range passed into the load() method.

Author:

Ronald B. Cemer

Field Summary

private int	charValue
private boolean	debug
private HashMap<Character,ArrayList<TrainingImage>>	dest
private DocumentScanner	documentScanner
private static Logger	LOG

Constructor Summary

TrainingImageLoader()

Method Summary

DocumentScanner	getDocumentScanner()
void	load(Image image, CharacterRange charRange, HashMap<Character,ArrayList<TrainingImage>> dest, String imageFilename)
void	load(String imageFilename, CharacterRange charRange, HashMap<Character,ArrayList<TrainingImage>> dest) Load an image containing training characters, break it up into characters, and build a training set.
void	processChar(PixelImage pixelImage, int x1, int y1, int x2, int y2, int rowY1, int rowY2)
void	setDebug(boolean debug)

Methods inherited from class net.sourceforge.javaocr.scanner.DocumentScannerListenerAdaptor

beginDocument, beginRow, endDocument, endRow, processSpace

Methods inherited from class java.lang.Object

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

Field Detail

charValue

private int charValue

dest

private HashMap<Character,ArrayList<TrainingImage>> dest

debug

private boolean debug

documentScanner

private DocumentScanner documentScanner

LOG

private static final Logger LOG

Constructor Detail

TrainingImageLoader

public TrainingImageLoader()

Method Detail

getDocumentScanner

public DocumentScanner getDocumentScanner()

Returns:

The DocumentScanner instance that is used to load the training images. This is useful if the caller wants to adjust some of the scanner's parameters.

load

public void load(String imageFilename,
                 CharacterRange charRange,
                 HashMap<Character,ArrayList<TrainingImage>> dest)
          throws IOException

Load an image containing training characters, break it up into characters, and build a training set. Each entry in the training set (a Map) has a key which is a Character object whose value is the character code. Each corresponding value in the Map is an ArrayList of one or more TrainingImage objects which contain images of the character represented in the key.

Parameters:

imageFilename - The filename of the image to load.

charRange - A CharacterRange object representing the range of characters which is contained in this image.

dest - A Map which gets loaded with the training data. Multiple calls to this method may be made with the same Map to populate it with the data from several training images.

Throws:

IOException

setDebug

public void setDebug(boolean debug)

load

public void load(Image image,
                 CharacterRange charRange,
                 HashMap<Character,ArrayList<TrainingImage>> dest,
                 String imageFilename)
          throws IOException

Throws:

IOException

processChar

public void processChar(PixelImage pixelImage,
                        int x1,
                        int y1,
                        int x2,
                        int y2,
                        int rowY1,
                        int rowY2)

Specified by:

processChar in interface DocumentScannerListener

Overrides:

processChar in class DocumentScannerListenerAdaptor