org.openimaj.image.text.extraction

Class LiuSamarabanduTextExtractorMultiscale

java.lang.Object

org.openimaj.image.text.extraction.TextExtractor<FImage>

org.openimaj.image.text.extraction.LiuSamarabanduTextExtractorMultiscale

All Implemented Interfaces:

ImageProcessor<FImage>, Processor<FImage>

@Reference(type=Inproceedings,
           author={"Liu, X.","Samarabandu, J."},
           title="Multiscale Edge-Based Text Extraction from Complex Images",
           year="2006",
           booktitle="Multimedia and Expo, 2006 IEEE International Conference on",
           pages={"1721 ","1724"},
           month="july",
           number="",
           volume="",
           customData={"keywords","multiscale edge-based text extraction;printed document image;scene text;text detection;document image processing;edge detection;feature extraction;text analysis;","doi","10.1109/ICME.2006.262882","ISSN",""})
public class LiuSamarabanduTextExtractorMultiscale
extends TextExtractor<FImage>

An implementation of the multiscale text extractor from MULTISCALE EDGE-BASED TEXT EXTRACTION FROM COMPLEX IMAGES; Xiaoqing Liu and Jagath Samarabandu The University of Western Ontario http://ieeexplore.ieee.org/xpl/freeabs_all.jsp?arnumber=4036951.

This multiscale text extractor uses a Gaussian pyramid to produce the multiscale feature vector. From this, the basic text extraction algorithm is used (see the LiuSamarabanduTextExtractorBasic implementation) on each image and the results combined using across-scale addition.

Author:

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

Created:

28 Jul 2011

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
class	LiuSamarabanduTextExtractorMultiscale.PyramidTextExtractor This is the main processor for this text extractor.

Constructor Summary

Constructors

Constructor and Description
LiuSamarabanduTextExtractorMultiscale()

Method Summary

Modifier and Type	Method and Description
protected void	forceWait() Helper method for debugging when viewing images
Map<Rectangle,FImage>	getTextRegions() Get the text regions that can be extracted from an image.
boolean	isDoubleSizePyramid() Whether the initial image in the pyramid is being double sized.
void	processImage(FImage image) Process an image.
void	setDoubleSizePyramid(boolean doubleSizePyramid) Set whether to double the size of the pyramid

Methods inherited from class org.openimaj.image.text.extraction.TextExtractor

getOCRProcessor, getText, getTextStrings, setOCRProcessor

Methods inherited from class java.lang.Object

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

Constructor Detail

LiuSamarabanduTextExtractorMultiscale

public LiuSamarabanduTextExtractorMultiscale()

Method Detail

forceWait

protected void forceWait()

Helper method for debugging when viewing images

processImage

public void processImage(FImage image)

Process an image. Implementing classes must alter the image passed in-place or assign the output to the input using Image.internalAssign(Image).

Parameters:

image - The image to process in place.

See Also:

ImageProcessor.processImage(org.openimaj.image.Image)

getTextRegions

public Map<Rectangle,FImage> getTextRegions()

Get the text regions that can be extracted from an image. The images in the values of the map need not be simply the extracted region that is bounded by the rectangular key (this can be done afterwards), but may be a representation that is as near to canonical as possible - that is, it may be warped or thresholded such that an OCR processor may have less trouble reading the text

Specified by:

getTextRegions in class TextExtractor<FImage>

Returns:

A map from bounding box in original image to a canonical representation of the text (may be warped or thresholded)

See Also:

TextExtractor.getTextRegions()

isDoubleSizePyramid

public boolean isDoubleSizePyramid()

Whether the initial image in the pyramid is being double sized.

Returns:

TRUE if the initial image is double sized.

setDoubleSizePyramid

public void setDoubleSizePyramid(boolean doubleSizePyramid)

Set whether to double the size of the pyramid

Parameters:

doubleSizePyramid - TRUE to double the size of the initial image.