We Keep Coding

JAI Tiff to JPEG convertion issue

I have an issue converting Tiff-Files to JPEGs with JAI. This is my Code:
ByteArrayOutputStream baos = new ByteArrayOutputStream();
TIFFDecodeParam param = null;
ImageDecoder dec = ImageCodec.createImageDecoder("tiff", new FileSeekableStream(inPath), param);
RenderedImage op = dec.decodeAsRenderedImage(0);
JPEGEncodeParam jpgparam = new JPEGEncodeParam();
jpgparam.setQuality(67);
ImageEncoder en = ImageCodec.createImageEncoder("jpeg", baos, jpgparam);
en.encode(op);
Mostly this code works fine, but with some Images, I got the following error:
java.lang.RuntimeException: Only 1, or 3-band byte data may be written.
at com.sun.media.jai.codecimpl.JPEGImageEncoder.encode(JPEGImageEncoder.java:142)
I cant find any related Problems over here and i have no idea how to fix it. The Images who throw this error have a high Resolution (9000 x 7000 or more) and are mostly scans of old pictures.
Image with this ColorModel works:
ColorModel:
#pixelBits = 24
numComponents = 3
color space = java.awt.color.ICC_ColorSpace#21981a50
transparency = 1 has alpha = false
isAlphaPre = false
This not:
ColorModel:
#pixelBits = 16
numComponents = 1
color space = java.awt.color.ICC_ColorSpace#88a30ad
transparency = 1 has alpha = false
isAlphaPre = false

I tried reading the JPEG standard, and it is not readily clear whether this is a limitation of the JPEG format or just the Encoder.
The encoder provide with java only encodes 1 or 3 byte bands, so in your case there are 16bit gray scale images. One way to solve this, as it appears you have done, is to save the image using a PNG encoder. It would not support the compression quality parameter.
The other way to handle this would be to save your image as an 8bit gray scale image.
I made a simple example to test this w/out JAI.
public static void main(String[] args) throws Exception{
BufferedImage img = new BufferedImage(256, 256, BufferedImage.TYPE_USHORT_GRAY);
Iterator<ImageWriter> writers = ImageIO.getImageWritersBySuffix("jpg");
while( writers.hasNext() ){
ImageWriter writer = writers.next();
ImageOutputStream ios = ImageIO.createImageOutputStream( new File("junk.jpg") );
writer.setOutput(ios);
writer.write(img);
}
}
The simplest way I can see to convert it is to create a new image and draw to it with a graphics.
BufferedImage img2 = new BufferedImage(img.getWidth(), img.getHeight(), BufferedImage.TYPE_BYTE_GRAY);
Graphics g = img2.getGraphics();
g.drawImage(img, 0, 0);
g.dispose();
Then img2 can be saved as JPG.

pdfbox - pdf increase size after converting to grayscale

I need to convert scanned PDF to grayscale PDF. I found 2 solutions for that.
First one is to just use renderImage
private void convertToGray() throws IOException {
File pdfFile = new File(PATH);
try (PDDocument originalPdf = PDDocument.load(pdfFile);
PDDocument doc = new PDDocument()) {
LOGGER.info("Current heap after loading file: {}", Runtime.getRuntime().totalMemory());
PDFRenderer pdfRenderer = new PDFRenderer(originalPdf);
for (int pageNum = 0; pageNum < originalPdf.getNumberOfPages(); pageNum++) {
// PDImageXObject pdImage = LosslessFactory.createFromImage(doc, bufferedImage);
BufferedImage grayImage = pdfRenderer.renderImageWithDPI(pageNum, 300F, ImageType.GRAY);
PDImageXObject pdImage = JPEGFactory.createFromImage(doc, grayImage);
float pageWight = originalPdf.getPage(pageNum).getMediaBox().getWidth();
float pageHeight = originalPdf.getPage(pageNum).getMediaBox().getHeight();
PDPage page = new PDPage(new PDRectangle(pageWight, pageHeight));
doc.addPage(page);
try (PDPageContentStream contentStream = new PDPageContentStream(doc, page)) {
contentStream.drawImage(pdImage, 0F, 0F, pageWight, pageHeight);
}
}
doc.save(NEW_PATH);
}
}
But this leads to increase size of the file (because some PDFs has less DPI than 300.
Second one is to just replace existing image with gray analog
private void convertByImageToGray() throws IOException {
File pdfFile = new File(PATH);
try (PDDocument document = PDDocument.load(pdfFile)) {
List<COSObject> objects = document.getDocument().getObjectsByType(COSName.IMAGE);
for (COSObject object : objects) {
LOGGER.info("Class: {}; {}", object.getClass(), object.toString());
}
for (int pageNum = 0; pageNum < document.getNumberOfPages(); pageNum++) {
PDPage page = document.getPage(pageNum);
replaceImage(document, page);
}
document.save(NEW_PATH);
}
}
private void replaceImage(PDDocument document, PDPage page) throws IOException {
PDResources resources = page.getResources();
Iterable<COSName> xObjectNames = resources.getXObjectNames();
if (xObjectNames != null) {
for (COSName xObjectName : xObjectNames) {
PDXObject object = resources.getXObject(xObjectName);
if (object instanceof PDImageXObject) {
PDImageXObject img1 = (PDImageXObject) object;
BufferedImage bufferedImage1 = img1.getImage();
BufferedImage grayBufferedImage = convertBufferedImageToGray(bufferedImage1);
// PDImageXObject grayImage = JPEGFactory.createFromImage(document, grayBufferedImage);
PDImageXObject grayImage = LosslessFactory.createFromImage(document, grayBufferedImage);
resources.put(xObjectName, grayImage);
}
}
}
}
private static BufferedImage convertBufferedImageToGray(BufferedImage sourceImg) {
ColorSpace cs = ColorSpace.getInstance(ColorSpace.CS_GRAY);
ColorConvertOp op = new ColorConvertOp(sourceImg.getColorModel().getColorSpace(), cs, null);
op.filter(sourceImg, sourceImg);
return sourceImg;
}
But still some files increase in size like 3 times (even they were already grayscale; interesting that int this case JPEGFactory produces larger files than LosslessFactory). All images in grayscale PDF have the same size as original ones. And I don't understand why.
Maybe there is a better way to make grayscale PDF with predictable size (except ghostscript)?
UPDATE: I've just realized that the issue is with creating PDF from image. It does not compress as well.
For example, I have dummy 1-page scan file that is less than 1 Mb. But if I get image from it (directly copying via Acrobat Reader to Paint, or via code above) it size is ~8-10 Mb depending on the method. And if I create new PDF from this image it's barely compressed. Here is example code:
File pdfFile = new File(FULL_FILE);
try (PDDocument document = PDDocument.load(pdfFile)) {
PDPage page = new PDPage();
document.addPage(page);
PDImageXObject pdImage = PDImageXObject.createFromFile("example.png", document);
try (PDPageContentStream contents = new PDPageContentStream(document, page)) {
contents.drawImage(pdImage, 0F, 0F);
}
document.save(FULL_FILE_NEW);
}

Yes LosslessFactory produces smaller files compared to JPEGFactory
In the below link there are different methods to try and achieve the same goal. Overall the best quality gray scale image was the one from Option 6, however this was by no means the fastest (I myself used Option 4). Comparisons are also provided for you to choose
This link contains possible ways to convert color images to black. It helped me a lot.
Let me know if it works for you and approve my answer if it helped.

Java : PDF page preview error, expected xref

I am trying to create a preview for PDF files which are created by Ballasamic Mockups. Around 50% of the time, I am not getting a preview and getting xref missing error. What am I doing wrong?
Error log :
com.sun.pdfview.PDFParseException: Expected 'xref' at start of table
at com.sun.pdfview.PDFFile.readTrailer(PDFFile.java:974)
at com.sun.pdfview.PDFFile.parseFile(PDFFile.java:1175)
at com.sun.pdfview.PDFFile.<init>(PDFFile.java:126)
at com.sun.pdfview.PDFFile.<init>(PDFFile.java:102)
Code :
private byte[] onlyCreatePdfPreview(String path, int attachId) {
try {
File file = new File(path);
RandomAccessFile raf = new RandomAccessFile(file, "r");
FileChannel channel = raf.getChannel();
ByteBuffer buf = channel.map(FileChannel.MapMode.READ_ONLY, 0, channel.size());
PDFFile pdffile = new com.sun.pdfview.PDFFile(buf);
PDFPage page = pdffile.getPage(0);
Rectangle rect = new Rectangle(0, 0,
(int) page.getBBox().getWidth(),
(int) page.getBBox().getHeight());
java.awt.Image img = page.getImage(
rect.width, rect.height, //width & height
rect, // clip rect
null, // null for the ImageObserver
true, // fill background with white
true // block until drawing is done
);
BufferedImage buffered = toBufferedImage(img);
buffered = Scalr.resize(buffered, Scalr.Method.ULTRA_QUALITY, 400, 250);
ByteArrayOutputStream baos = new ByteArrayOutputStream();
ImageIO.write(buffered, "png", baos);
baos.flush();
return baos.toByteArray();
} catch (Exception e) {
e.printStackTrace();
}
}
What am I doing wrong? Thank you.
Final Working code
try {
String pdfPath = zipLocation + String.valueOf(new BigInteger(130, random).toString(32));
PdfReader reader = new PdfReader(path);
PdfStamper pdfStamper = new PdfStamper(reader,new FileOutputStream(pdfPath));
pdfStamper.getWriter().setPdfVersion(PdfWriter.PDF_VERSION_1_4);
pdfStamper.close();
reader.close();
RandomAccessFile raf = new RandomAccessFile(pdfPath, "r");
FileChannel channel = raf.getChannel();
ByteBuffer buf = channel.map(FileChannel.MapMode.READ_ONLY, 0, channel.size());
PDFFile pdffile = new com.sun.pdfview.PDFFile(buf);
PDFPage page = pdffile.getPage(0);
Rectangle rect = new Rectangle(0, 0,
(int) page.getBBox().getWidth(),
(int) page.getBBox().getHeight());
java.awt.Image img = page.getImage(
rect.width, rect.height, //width & height
rect, // clip rect
null, // null for the ImageObserver
true, // fill background with white
true // block until drawing is done
);
BufferedImage buffered = toBufferedImage(img);
buffered = Scalr.resize(buffered, Scalr.Method.ULTRA_QUALITY, 400, 250);
ByteArrayOutputStream baos = new ByteArrayOutputStream();
ImageIO.write(buffered, "png", baos);
baos.flush();
return baos.toByteArray();
}//catch block

(This answer collects information from comments to the question which eventually led to a solution.)
Apparently com.sun.pdfview.PDFFile expects the cross references to start with xref. But this expectation only makes sense for PDFs following a PDF Reference up to revision 3 (version 1.4) published November 2001; PDFs following a later Reference or even the ISO 32000 standard (part 1 or 2) have the choice of using a cross reference stream (starting with an object number) instead of a cross reference table (starting with xref).
Thus, one should switch to using software following newer specifications than some more than 15 years old version or one has to convert one's PDFs to follow the old specifications, at least on the surface.
One can convert manually (e.g. using Adobe Acrobat) or automatized (e.g. using iText). (These examples software products really are only examples, other products can also be used for this task.)
If using a current iText 5 version, the conversion looks like this:
PdfReader reader = new PdfReader(SOURCE);
PdfStamper stamper = new PdfStamper(reader, DEST);
stamper.getWriter().setPdfVersion(PdfWriter.PDF_VERSION_1_4);
stamper.close();
reader.close();
And one has to take care that if SOURCE is a file name or a random access file, DEST must not be a file output stream to the same file. Otherwise the original file is truncated before the PdfStamper had a chance to copy it all to its output.

BufferedImage color saturation

I'm writing a simple scanning application using jfreesane and Apache PDFBox.
Here is the scanning code:
InetAddress address = InetAddress.getByName("192.168.0.17");
SaneSession session = SaneSession.withRemoteSane(address);
List<SaneDevice> devices = session.listDevices();
SaneDevice device = devices.get(0);
device.open();
device.getOption("resolution").setIntegerValue(300);
BufferedImage bimg = device.acquireImage();
File file = new File("test_scan.png");
ImageIO.write(bimg, "png", file);
device.close();
And making PDF:
PDDocument document = new PDDocument();
float width = bimg.getWidth();
float height = bimg.getHeight();
PDPage page = new PDPage(new PDRectangle(width, height));
document.addPage(page);
PDImageXObject pdimg = LosslessFactory.createFromImage(document, bimg);
PDPageContentStream stream = new PDPageContentStream(document, page, PDPageContentStream.AppendMode.APPEND, true);
stream.drawImage(pdimg, 0, 0);
stream.close();
document.save(filename);
document.close();
And here is the result:
As you can see the PDF image is more "pale" (saturation? - sorry, I'm not good at color theory and don't know how to name it correctly).
What I have found out:
Printing BufferedImage to JLabel using JLabel(new ImageIcon(bimg))
constructor produces the same result as with PDF ("pale" colors)
so I guess PDFBox is not the reason.
Changing scanning resolution -
no effect.
bimg.getTransparency() returns 1 (OPAQUE)
bimg.getType() returns 0 (TYPE_CUSTOM)
PNG file:
http://s000.tinyupload.com/index.php?file_id=95648202713651192395
PDF file
http://s000.tinyupload.com/index.php?file_id=90369236997064329368

There was an issue in JFreeSane with colorspaces, it was fixed in version 0.97:
https://github.com/sjamesr/jfreesane/releases/tag/jfreesane-0.97

Replacing images with same resource in PDFBox

I have a pdf containing 2 blank images. I need to replace both the images with 2 separate images using PDFBox. The problem is, both the blank images appear to have the same resource. So, if I replace one, the other one is replaced with the same image as well.
I followed this example and tried overriding the processOperator() method and replaced the images based on the imageHeight. However, it still ends up replacing both the images with the same image. This is my code thus far:
protected void processOperator( PDFOperator operator, List arguments ) throws IOException
{
String operation = operator.getOperation();
if( INVOKE_OPERATOR.equals(operation) )
{
COSName objectName = (COSName)arguments.get( 0 );
Map<String, PDXObject> xobjects = getResources().getXObjects();
PDXObject xobject = (PDXObject)xobjects.get( objectName.getName() );
if( xobject instanceof PDXObjectImage )
{
PDXObjectImage blankImage = (PDXObjectImage)xobject;
int imageWidth = blankImage.getWidth();
int imageHeight = blankImage.getHeight();
System.out.println("Image width >>> "+imageWidth+" height >>>> "+imageHeight);
// Check if it is blank image 1 based on height
if(imageHeight < 480){
File logo = new File("abc.jpg");
BufferedImage bufferedImage = ImageIO.read(logo);
ByteArrayOutputStream baos = new ByteArrayOutputStream();
ImageIO.write( bufferedImage, "jpg", baos );
baos.flush();
byte[] logoImageInBytes = baos.toByteArray();
baos.close();
// label will be used to replace the blank image
label = logoImageInBytes;
}
BufferedImage img = ImageIO.read(new ByteArrayInputStream(label));
BufferedImage resizedImage = Scalr.resize(img, Scalr.Method.BALANCED, Scalr.Mode.FIT_EXACT, img.getWidth(), img.getHeight());
ByteArrayOutputStream baos = new ByteArrayOutputStream();
ImageIO.write(resizedImage, "jpg", baos);
// Replace empty image in template with the image generated from shipping label byte array
PDXObjectImage validImage = new PDJpeg(doc, new ByteArrayInputStream(baos.toByteArray()));
blankImage.getCOSStream().replaceWithStream(validImage.getCOSStream());
}
Now, when I remove the if block which checks if (imageHeight < 480), it prints the imageHeight as 30 and 470 for the blank images. However, when I add the if block, it prints the imageHeight as 480 and 1500 and never goes inside the if block because of which both the blank images end up getting replaced by the same image.
What's going on here? I'm new to PDFBox, so I am unsure if my code is correct.

While first thinking about a generic way to actually replace the existing Image by the new Images, I agree with #TilmanHausherr that a more simple solution would be to simply add an extra content stream with two images in the size / position you need covering the existing Image.
This approach is easier to implement (even generically) and less error-prone than actual replacement.
In a generic solution we do not have the Image positions beforehand. To determine them, we can use this helper class (which essentially is a rip-off of the PDFBox example PrintImageLocations):
public class ImageLocator extends PDFStreamEngine
{
private static final String INVOKE_OPERATOR = "Do";
public ImageLocator() throws IOException
{
super(ResourceLoader.loadProperties("org/apache/pdfbox/resources/PDFTextStripper.properties", true));
}
public List<ImageLocation> getLocations()
{
return new ArrayList<ImageLocation>(locations);
}
protected void processOperator(PDFOperator operator, List<COSBase> arguments) throws IOException
{
String operation = operator.getOperation();
if (INVOKE_OPERATOR.equals(operation))
{
COSName objectName = (COSName) arguments.get(0);
Map<String, PDXObject> xobjects = getResources().getXObjects();
PDXObject xobject = (PDXObject) xobjects.get(objectName.getName());
if (xobject instanceof PDXObjectImage)
{
PDXObjectImage image = (PDXObjectImage) xobject;
PDPage page = getCurrentPage();
Matrix matrix = getGraphicsState().getCurrentTransformationMatrix();
locations.add(new ImageLocation(page, matrix, image));
}
else if (xobject instanceof PDXObjectForm)
{
// save the graphics state
getGraphicsStack().push((PDGraphicsState) getGraphicsState().clone());
PDPage page = getCurrentPage();
PDXObjectForm form = (PDXObjectForm) xobject;
COSStream invoke = (COSStream) form.getCOSObject();
PDResources pdResources = form.getResources();
if (pdResources == null)
{
pdResources = page.findResources();
}
// if there is an optional form matrix, we have to
// map the form space to the user space
Matrix matrix = form.getMatrix();
if (matrix != null)
{
Matrix xobjectCTM = matrix.multiply(getGraphicsState().getCurrentTransformationMatrix());
getGraphicsState().setCurrentTransformationMatrix(xobjectCTM);
}
processSubStream(page, pdResources, invoke);
// restore the graphics state
setGraphicsState((PDGraphicsState) getGraphicsStack().pop());
}
}
else
{
super.processOperator(operator, arguments);
}
}
public class ImageLocation
{
public ImageLocation(PDPage page, Matrix matrix, PDXObjectImage image)
{
this.page = page;
this.matrix = matrix;
this.image = image;
}
public PDPage getPage()
{
return page;
}
public Matrix getMatrix()
{
return matrix;
}
public PDXObjectImage getImage()
{
return image;
}
final PDPage page;
final Matrix matrix;
final PDXObjectImage image;
}
final List<ImageLocation> locations = new ArrayList<ImageLocation>();
}
(ImageLocator.java)
In contrast to the example class this helper stores the locations in a list instead of printing them.
We now can cover existing images using code like this:
try ( InputStream resource = getClass().getResourceAsStream("sample.pdf");
InputStream left = getClass().getResourceAsStream("left.png");
InputStream right = getClass().getResourceAsStream("right.png");
PDDocument document = PDDocument.load(resource) )
{
if (document.isEncrypted())
{
document.decrypt("");
}
PDJpeg leftImage = new PDJpeg(document, ImageIO.read(left));
PDJpeg rightImage = new PDJpeg(document, ImageIO.read(right));
// Locate images
ImageLocator locator = new ImageLocator();
List<?> allPages = document.getDocumentCatalog().getAllPages();
for (int i = 0; i < allPages.size(); i++)
{
PDPage page = (PDPage) allPages.get(i);
locator.processStream(page, page.findResources(), page.getContents().getStream());
}
// cover images
for (ImageLocation location : locator.getLocations())
{
// Decide on a replacement
PDRectangle cropBox = location.getPage().findCropBox();
float center = (cropBox.getLowerLeftX() + cropBox.getUpperRightX()) / 2.0f;
PDJpeg image = location.getMatrix().getXPosition() < center ? leftImage : rightImage;
AffineTransform transform = location.getMatrix().createAffineTransform();
PDPageContentStream content = new PDPageContentStream(document, location.getPage(), true, false, true);
content.drawXObject(image, transform);
content.close();
}
document.save(new File(RESULT_FOLDER, "sample-changed.pdf"));
}
(OverwriteImage)
This sample covers all images on the left half of their respective page with left.png and all others with right.png.

I have no implementation or example, but I want to illustrate you a possible way to do what you want by the following steps:
Since you need 2 Images (lets tell them imageA and imageB) in the pdf instead of 1 (which is the blank one). You have to add both of them to the pdf.
save the file temporary - optional, it could work without rewriting the pdf
reopen the file - optional, if you don't need step 2, you also don't need this step
Then replace the blank image with imageA or imageB
Remove the blank image from the pdf
Save the pdf