I have a few PDF files which suffer the problem shown in the image:

Acceptable embedded font, UNacceptable rendering (the blue part is mouse-selected, the right part in black is how the entire document looks!)

I tried /prepressing and even -dSUBSTFONTing them with ghostscript but it didn't help, and I'm clueless now.

Any suggestions? ... Thanks.

  • 2
    Notice the jagged margins of the letters of the black text? That means it's a raster image. The fact you can select it means your PDF also has an OCR layer. That's how PDF works, there isn't anything to fix there (except perhaps the alignment of the OCR layer; blame the author of the PDF for that, not your tools). – Satō Katsura Jul 4 '16 at 9:58
  • Thank you. Then is there a way to use the OCR generated version only, discarding the existing file? I think that could be more readable even if the OCR does it's job partially. I appreciate further comments @SatoKatsura :-) – Haix64 Jul 4 '16 at 17:46
  • Some mobile PDF readers can show you the OCR layer. I'm not aware of any desktop reader that can do the same. You can in principle extract the text (with, say, pdftotext from the poppler package), but you'd lose any information about sizes and positioning, and spacing between words will be occasionally wrong (words are essentially determined heuristically). Depending on the quality of OCR, the result may or may not be usable. – Satō Katsura Jul 4 '16 at 18:35

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.