Hi There,
I've been trying to highlight keywords in a tiff converted to OCR PDF and it does not work as expected. It does add a patch above the text, however entire point of highlighting is not helping since I can't see the text that it highlighted. I've attached the pdf files so that you guys understand what I am trying to say.
Any help on this will be appreciated.
Kind Regards,
Vasanth
Hi Vasanth,
Not exactly. The tiff image was converted to OCR PDF and I used aspose.net pdf to highlight the text with yellow color programmatically. That's when it adds the highlight color and hides the text beneath.
Uploaded the tiff image as zip. Please see attachment.
Hi Vasanth,
// instantiate PDF object<o:p></o:p>
Pdf pdf = new Pdf();
// create Section instance
Section CurrentPage = pdf.Sections.Add();
// create Image object
Aspose.Pdf.Generator.Image img = new Aspose.Pdf.Generator.Image();
CurrentPage.Paragraphs.Add(img);
// set source image file path
img.ImageInfo.File = @"C:\test.tiff";
// set the image file type
img.ImageInfo.ImageFileType = ImageFileType.Tiff;
// add all frames of tiff image to PDF file
img.ImageInfo.TiffFrame = -1;
MemoryStream stream = new MemoryStream();
// save the output image
pdf.Save(stream);
// create content editor object
Aspose.Pdf.Facades.PdfContentEditor editor = new PdfContentEditor();
// bind the source PDF file
editor.BindPdf(stream);
// create highlight annotation
editor.CreateMarkup(new System.Drawing.Rectangle(303, 645, 30, 10), "", 0, 1, System.Drawing.Color.Yellow);
// save updated file
editor.Save(@"C:\test_Highlight.pdf");
// close stream object
stream.Close();
Appreciate your help. However at this point the pdf is not searchable for text anymore right? Can this be overcome? Also you have hardcoded the highlight using co-ordinates, in my case I am supposed to search for keywords in multiple pages and highlight the words and possibly the text should be searchable in the PDF. I know its a lot of expectation, however that would be our true solution to the problem we are facing at work.
Any help on this will be appreciated.
Hi Vasanth,
Thanks for sharing the details.
As the source TIFF image is converted to PDF format, so the text is not searchable in PDF file. However in order to accomplish your requirement, you may first try performing OCR on source TIFF image and then try adding the text to PDF file. Once the text is added to PDF file, you can easily search the text and highlight the text fragments.
In my earlier post, I have hardcoded the coordinates of text inside PDF file because the text cannot be searched inside the PDF. However once the text is converted to PDF file, you can search the text inside PDF and perform required operations. For further details, please visit
In the event of any further query, please feel free to contact.
I've looked at the links you mentioned before starting this support thread.
In my initial post, that's exactly what I did to highlight the text and that is when the textfragment highlight masked the original text. I attached the ocr pdf in the initial post and I am attaching it now so that you can take a look at it.
If you are able to highlight the text in this pdf without masking the text, please let me know how you did it. "test" is the word I highlighted using regex so that it would highlight "Test" and "test".
Again appreciate your help.
Is there a way we could do WebEx or call? Since we have had quite a few back and forth on this discussion. I wanted to make sure you understood the issue we are having for which we are trying to implement a solution.
Hi Vasanth,
//open document<o:p></o:p>
Document pdfDocument = new Document("c:/pdftest/test_from-graphic_ocr (1).pdf");
//create TextAbsorber object to find all instances of the input search phrase
TextFragmentAbsorber textFragmentAbsorber = new TextFragmentAbsorber("Test");
//accept the absorber for all the pages
pdfDocument.Pages[1].Accept(textFragmentAbsorber);
//get the extracted text fragments
TextFragmentCollection textFragmentCollection = textFragmentAbsorber.TextFragments;
// create Rectangle object to hold TextFragment rectangular coordinates
System.Drawing.Rectangle text_rect = new System.Drawing.Rectangle();
//loop through the fragments
foreach (TextFragment textFragment in textFragmentCollection)
{
foreach (TextSegment textSegment in textFragment.Segments)
{
// print text segment
Console.WriteLine("Text : {0} ", textSegment.Text);
Console.WriteLine("XIndent : {0} ",textSegment.Position.XIndent);
Console.WriteLine("YIndent : {0} ",textSegment.Position.YIndent);
// create Rectangle object based on TextFragment rectangle coordinates
text_rect = new System.Drawing.Rectangle((int)textFragment.Rectangle.LLX, (int)textFragment.Rectangle.LLY, (int)textFragment.Rectangle.URX, (int)textFragment.Rectangle.URY);
// print Rectangular coordinates of text fragment
Console.WriteLine("LLX = "+(int)textFragment.Rectangle.LLX + " LLY = " + (int)textFragment.Rectangle.LLY +" URX = " + (int)textFragment.Rectangle.URX +" URY = "+ (int)textFragment.Rectangle.URY);
}
}
// create content editor object
Aspose.Pdf.Facades.PdfContentEditor editor = new PdfContentEditor();
// bind the source PDF file
editor.BindPdf(pdfDocument);
// create markup
editor.CreateMarkup(text_rect, "", 0, 1, System.Drawing.Color.Yellow);
// save updated file
editor.Save(@“C:\pdftest\test_Highlight.pdf”);
skvasant:
Is there a way we could do WebEx or call? Since we have had quite a few back and forth on this discussion. I wanted to make sure you understood the issue we are having for which we are trying to implement a solution.
The issues you have found earlier (filed as PDFNEWNET-36059) have been fixed in Aspose.Pdf for .NET 8.7.0.
Hi Vasanth,