org.openimaj.image.text.extraction

Class TextExtractor<T extends Image<?,T>>

java.lang.Object

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

Type Parameters:

T - The type of Image

All Implemented Interfaces:

ImageProcessor<T>, Processor<T>

Direct Known Subclasses:

LiuSamarabanduTextExtractorBasic, LiuSamarabanduTextExtractorMultiscale

public abstract class TextExtractor<T extends Image<?,T>>
extends Object
implements ImageProcessor<T>

An interface for classes that are able to extract text from images. The single method allows the retrieval of the text mapped to the bounding boxes of the text within the image.

Note that this is an ImageProcessor extension so that the TextExtractor should process the image prior to the getTextRegions() method being called.

This class will deal with the processing of extracted text regions with the OCR processor. Use setOCRProcessor(OCRProcessor) to choose with OCR processor will be used on the extracted regions.

Author:

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

Constructor Summary

Constructors

Constructor and Description
TextExtractor()

Method Summary

Methods

Modifier and Type	Method and Description
OCRProcessor<T>	getOCRProcessor() Return the OCR processor being used to extract text from the image.
Map<Rectangle,IndependentPair<T,String>>	getText() Get text that can be extracted from an image.
abstract Map<Rectangle,T>	getTextRegions() Get the text regions that can be extracted from an image.
List<String>	getTextStrings() If you're not interested in where the strings are located in the image you can use this method to simply get a list of extracted strings.
void	setOCRProcessor(OCRProcessor<T> ocr) For the text regions that are extracted to be associated with textual representations of the text regions, an OCR processor must be used.

Methods inherited from class java.lang.Object

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

Methods inherited from interface org.openimaj.image.processor.ImageProcessor

processImage

Constructor Detail

TextExtractor

public TextExtractor()

Method Detail

getTextRegions

public abstract Map<Rectangle,T> 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

Returns:

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

getText

public Map<Rectangle,IndependentPair<T,String>> getText()

Get text that can be extracted from an image. The map should map a bounding box within the processed image to a pair of extracted image vs. text string. The extracted image may not necessarily be the region of interest which the rectangle bounds; it can be as close to a canonical representation of the text as possible such that an OCR would have less difficulty in classifying the text. For example, the image may be thresholded or warped such that the text is straight.

Returns:

A map of bounding box to a pair of image and text string

getTextStrings

public List<String> getTextStrings()

If you're not interested in where the strings are located in the image you can use this method to simply get a list of extracted strings.

Returns:

A List of strings extracted from the image.

setOCRProcessor

public void setOCRProcessor(OCRProcessor<T> ocr)

For the text regions that are extracted to be associated with textual representations of the text regions, an OCR processor must be used. Use this function to choose which OCR processor is used to extract read text regions.

Parameters:

ocr - The OCRProcessor to use

getOCRProcessor

public OCRProcessor<T> getOCRProcessor()

Return the OCR processor being used to extract text from the image.

Returns:

The OCRProcessor