When I open an existing PDF, the OCR interpretation creates erroneous text. I just want to view the existing PDF without the erroneous text.
There is no way to prevent LibreOffice Draw from using OCR to recognise text in a PDF. I have found that the best solution is to open the PDF in LibreOffice Draw, then go page by page to click-drag a select box around the text - taking care to avoid the page margins of the page - and hitting delete. This preserves the original image, which you can then crop (to get rid of dark margins) and add text or images or print as is.
1 Like
Thank you so much for this great idea.
There is no way to prevent LibreOffice Draw from using OCR to recognise text in a PDF.
LOL. What a astonishing nonsense!
There’s no way to make LibreOffice Draw to use OCR to recognize text in a PDF. There’s no OCR built into LibreOffice. There’s no code in it to use any external OCR. There’s absolutely no magic like that. LibreOffice can only import the text objects that already exists in the PDF (so if a PDF imports some text boxes atop of a raster image, then the PDF already had been created from a pre-processed source, and LibreOffice only imported the existing objects).
1 Like
If you want to view a PDF, open it with a PDF Viewer and not with LibreOffice. I never got OCR problems when viewing PDF’s, or was your PDF created from a scan?
Edit (on comment): LibreOffice is not a PDF editor, nor a PDF viewer. If you scan a document, it’s probably better to scan to a text editor, in Windows that’d be Notepad or Wordpad, I’m not sure what’d work on Linux. You can then edit any OCR errors, either in the text editor or in LibreOffice. Scanning to PDF only makes sense if you don’t want to edit the file afterwards (for instance when it’s a picture).
1 Like
I scanned a document to a pdf file on my computer and then opened the pdf file in LibreOffice Draw and Writer.
In any case LIbreOffice doesn’t do any OCR interpretation, so such erroneous text must be in the PDF.
1 Like