pdfbox-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Olaf Drümmer <olafl...@callassoftware.com>
Subject Re: how to create structure for an existing PDF document
Date Thu, 12 Mar 2015 13:22:16 GMT
Hi Klaus,

if you are creating tagged PDF for accessibility purposes, and your PDFs consist of one image
per page, the use of alternate descriptions is questionable (unless each scanned page is actually
perceived as an image, and not as a scanned image of text). Where images from scanned pages
consist of text, it is instead more suitable to use OCR and place an invisible text layer
onto each page. The text in this invisible text layer would then be organised into a suitable
tagging structure.

The standard applicable to accessible tagged PDF is PDF/UA (ISO 14289-1). The PDF Association's
website (www.pdfa.org) offers an introductory booklet (as PDF for free of charge download),
and if you are interested in the more technical details, it is worthwhile looking at the Matterhorn
Protocol, which explains in detail about the rules an accessible PDF has to comply with if
it aims to  conform with the PDF/UA standard.

Olaf

PS: I have no clue about the PDFBox API, so I cannot be of help in this regard.


On 12 Mar 2015, at 13:20, "Henning, Klaus" <KHenning@eitco.de> wrote:

> Hi Olaf,
> 
> I want to create structure in the sense of a tagged PDF, so I can add or modify the alternate
description of all images in an PDF document. 
> The only way I found to add or modify an the alternate description of an image is via
the structreelement of it.
> 
> structureElement.setAlternateDescription("NEW ALTERNATE DESCRIPTION");
> 
> As described in my previous mail I want to recognitze the text in the image via ocr and
write the returned text to the alternate description of the image.
> 
> 
> Klaus
> 
> 
> -----Ursprüngliche Nachricht-----
> Von: Olaf Drümmer [mailto:olaflist@callassoftware.com] 
> Gesendet: Donnerstag, 12. März 2015 12:33
> An: users@pdfbox.apache.org
> Cc: Olaf Drümmer
> Betreff: Re: how to create structure for an existing PDF document
> 
> Hi Klaus,
> 
> what kind of structure do you wish to create? Structure in the sense of tagged PDF, or
just some logical structure, and if so, for what purposes?
> 
> Olaf
> 
> 
> On 12 Mar 2015, at 11:54, "Henning, Klaus" <KHenning@eitco.de> wrote:
> 
>> Hi,
>> 
>> we want to create the structure to an existing PDF document. We have PDF documents
from a scanner which contains Images but no structure.
>> We want to implement a program to create the structure so we can add AlternateDescriptions
to the images based on tesaract ocr recognition.
>> 
>> Our first approach creates a structure but the structure seems to be 
>> incomplete when checking it with adobe acrobat. We can't find any hints in the pdfbox
examples or documentation how to do this.
>> 
>> Our Code snippet:
>> 
>>            try {
>>                   PDDocument document = PDDocument.load("test.pdf");
>>                   PDDocumentCatalog documentCatalog = 
>> document.getDocumentCatalog();
>> 
>>                   PDStructureTreeRoot treeRoot = 
>> document.getDocumentCatalog().getStructureTreeRoot();
>> 
>>                   if(treeRoot == null){
>>                          COSDictionary cosDictionary = documentCatalog.getCOSDictionary();
>>                          PDStructureTreeRoot newTreeRoot = new 
>> PDStructureTreeRoot();
>> 
>>                          //iterate over pages
>>                          List<?> pages = documentCatalog.getAllPages();
>>                          for (Object object : pages) {
>>                                 PDPage page = (PDPage) object;
>>                                 Map<String,PDXObject> mapObjects = page.getResources().getXObjects();
>>                                 for (PDXObject pdxObject : mapObjects.values()) {
>>                                       if(pdxObject instanceof PDXObjectImage){
>>                                              PDXObjectImage objectImage = (PDXObjectImage)pdxObject;
>>                                              //new SturctureElement for the image
>>                                              PDStructureElement structureElement
= new PDStructureElement(StandardStructureTypes.Figure,newTreeRoot);
>>                                              PDMarkedContent markedContent = new
PDMarkedContent(COSName.IMAGE,  new COSDictionary());
>>                                              markedContent.addXObject(objectImage);
>>                                              structureElement.appendKid(markedContent);
>>                                              structureElement.setAlternateDescription("NEW
ALTERNATE DESCRIPTION");
>>                                              newTreeRoot.appendKid(structureElement);
>>                                       }
>>                                 }
>>                          }
>> 
>>                          documentCatalog.setStructureTreeRoot(newTreeRoot);
>>                          treeRoot = documentCatalog.getStructureTreeRoot();
>>                   }
>> 
>>                   document.save("testWithTree.pdf");
>>                   document.close();
>>            }
>>            catch (IOException e) {
>>                   e.printStackTrace();
>>            }
>>            catch (COSVisitorException e) {
>>                   e.printStackTrace();
>>            }
>> 
>> Can someone help us her?
>> 
>> Best regards,
>> 
>> Klaus Henning
>> 
>> 
>> ______________________________________________________________________
>> This email has been scanned by the Symantec Email Security.cloud service.
>> For more information please visit http://www.symanteccloud.com 
>> ______________________________________________________________________
> 
> 
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: users-unsubscribe@pdfbox.apache.org
> For additional commands, e-mail: users-help@pdfbox.apache.org
> 
> ______________________________________________________________________
> This email has been scanned by the Symantec Email Security.cloud service.
> For more information please visit http://www.symanteccloud.com ______________________________________________________________________
> 
> ______________________________________________________________________
> This email has been scanned by the Symantec Email Security.cloud service.
> For more information please visit http://www.symanteccloud.com
> ______________________________________________________________________
> 
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: users-unsubscribe@pdfbox.apache.org
> For additional commands, e-mail: users-help@pdfbox.apache.org
> 


---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@pdfbox.apache.org
For additional commands, e-mail: users-help@pdfbox.apache.org


Mime
View raw message