public/docs
1
0
Fork 0
mirror of https://github.com/sismics/docs.git synced 2025-12-13 09:46:17 +00:00
Code Issues Releases Wiki Activity

Closes #59: Use TwelveMonkeys' ImageIO plugin for JPEG

Browse Source
This commit is contained in:
jendib
2016-01-24 15:44:40 +01:00
parent 456fc5b991
commit 3172a5f216
9 changed files with 84 additions and 118 deletions
Download Patch File Download Diff File Expand all files Collapse all files

25
docs-core/pom.xml
View File

@@ -113,11 +113,6 @@
<artifactId>bcprov-jdk15on</artifactId>
</dependency>
<dependency>
<groupId>com.levigo.jbig2</groupId>
<artifactId>levigo-jbig2-imageio</artifactId>
</dependency>
<dependency>
<groupId>fr.opensagres.xdocreport</groupId>
<artifactId>org.odftoolkit.odfdom.converter.pdf</artifactId>
@@ -127,16 +122,26 @@
<groupId>fr.opensagres.xdocreport</groupId>
<artifactId>org.apache.poi.xwpf.converter.pdf</artifactId>
</dependency>
<!-- OCR dependencies -->
<dependency>
<groupId>jna</groupId>
<groupId>net.java.dev.jna</groupId>
<artifactId>jna</artifactId>
</dependency>
<!-- ImageIO plugins -->
<dependency>
<groupId>com.levigo.jbig2</groupId>
<artifactId>levigo-jbig2-imageio</artifactId>
</dependency>
<dependency>
<groupId>com.twelvemonkeys.imageio</groupId>
<artifactId>imageio-jpeg</artifactId>
</dependency>
<dependency>
<groupId>jai</groupId>
<artifactId>imageio</artifactId>
<groupId>com.github.jai-imageio</groupId>
<artifactId>jai-imageio-core</artifactId>
</dependency>
<!-- Test dependencies -->

26
docs-core/src/main/java/com/sismics/tess4j/ImageIOHelper.java
View File

@@ -38,9 +38,9 @@ import javax.imageio.stream.ImageOutputStream;
import org.w3c.dom.NodeList;
import com.sun.media.imageio.plugins.tiff.TIFFImageWriteParam;
import com.sun.media.imageioimpl.plugins.tiff.TIFFImageReaderSpi;
import com.sun.media.imageioimpl.plugins.tiff.TIFFImageWriterSpi;
import com.github.jaiimageio.impl.plugins.tiff.TIFFImageReaderSpi;
import com.github.jaiimageio.impl.plugins.tiff.TIFFImageWriterSpi;
import com.github.jaiimageio.plugins.tiff.TIFFImageWriteParam;
public class ImageIOHelper {
@@ -51,26 +51,26 @@ public class ImageIOHelper {
* Gets pixel data of an
* <code>IIOImage</code> object.
*
* @param image an
* @param oimage an
* <code>IIOImage</code> object
* @return a byte buffer of pixel data
* @throws Exception
*/
public static ByteBuffer getImageByteBuffer(IIOImage image) throws IOException {
//Set up the writeParam
TIFFImageWriteParam tiffWriteParam = new TIFFImageWriteParam(Locale.US);
tiffWriteParam.setCompressionMode(ImageWriteParam.MODE_DISABLED);
//Get tif writer and set output to file
public static ByteBuffer getImageByteBuffer(BufferedImage oimage) throws IOException {
// Get tif writer and set output to file
ImageWriter writer = new TIFFImageWriterSpi().createWriterInstance();
//Get the stream metadata
// Set up the writeParam
// We are using the old JAI ImageIO plugin, because for some reason, OCR don't work with TwelveMonkeys' plugin
ImageWriteParam tiffWriteParam = new TIFFImageWriteParam(Locale.US);
tiffWriteParam.setCompressionMode(ImageWriteParam.MODE_DISABLED);
// Get the stream metadata
IIOMetadata streamMetadata = writer.getDefaultStreamMetadata(tiffWriteParam);
ByteArrayOutputStream outputStream = new ByteArrayOutputStream();
ImageOutputStream ios = ImageIO.createImageOutputStream(outputStream);
writer.setOutput(ios);
writer.write(streamMetadata, new IIOImage(image.getRenderedImage(), null, null), tiffWriteParam);
writer.write(streamMetadata, new IIOImage(oimage, null, null), tiffWriteParam);
writer.dispose();
// Read the writed image

17
docs-core/src/main/java/com/sismics/tess4j/Tesseract.java
View File

@@ -17,7 +17,6 @@ package com.sismics.tess4j;
import java.awt.Rectangle;
import java.awt.image.BufferedImage;
import java.awt.image.RenderedImage;
import java.io.IOException;
import java.nio.ByteBuffer;
import java.util.ArrayList;
@@ -25,8 +24,6 @@ import java.util.Enumeration;
import java.util.List;
import java.util.Properties;
import javax.imageio.IIOImage;
import com.sun.jna.Pointer;
/**
@@ -169,9 +166,8 @@ public class Tesseract {
* @throws TesseractException
*/
public String doOCR(BufferedImage bi, Rectangle rect) throws TesseractException {
IIOImage oimage = new IIOImage(bi, null, null);
List<IIOImage> imageList = new ArrayList<IIOImage>();
imageList.add(oimage);
List<BufferedImage> imageList = new ArrayList<BufferedImage>();
imageList.add(bi);
return doOCR(imageList, rect);
}
@@ -179,23 +175,22 @@ public class Tesseract {
* Performs OCR operation.
*
* @param imageList a list of
* <code>IIOImage</code> objects
* <code>BufferedImage</code> objects
* @param rect the bounding rectangle defines the region of the image to be
* recognized. A rectangle of zero dimension or
* <code>null</code> indicates the whole image.
* @return the recognized text
* @throws TesseractException
*/
public String doOCR(List<IIOImage> imageList, Rectangle rect) throws TesseractException {
public String doOCR(List<BufferedImage> imageList, Rectangle rect) throws TesseractException {
StringBuilder sb = new StringBuilder();
pageNum = 0;
for (IIOImage oimage : imageList) {
for (BufferedImage oimage : imageList) {
pageNum++;
try {
ByteBuffer buf = ImageIOHelper.getImageByteBuffer(oimage);
RenderedImage ri = oimage.getRenderedImage();
String pageText = doOCR(ri.getWidth(), ri.getHeight(), buf, rect, ri.getColorModel().getPixelSize());
String pageText = doOCR(oimage.getWidth(), oimage.getHeight(), buf, rect, oimage.getColorModel().getPixelSize());
sb.append(pageText);
} catch (IOException ioe) {
//skip the problematic image