finding keywords in PDFs/scans

Hello everyone, hopefully this will be solved quickly and people can get back to regular watmming.

SITUATION/CONTEXT: I have loads of pdfs, most of which I've had scanned, for my research. As I can't command-F these to find keywords I was wondering if there's another way (software?) to achieve this.

thanks in advance

look into OCR

looking into it now, thank you

sorry couldn't point you to any specific software. someone here is bound to know what's a good one

Google Docs has OCR.

i just use the ocr/text-recognition in acrobat x.

it's fine.

though you def need the full version so.

work on that.

Yes, any version of Adobe Acrobat Professional will have OCR capabilities, but keep in mind it will depend on what quality you scanned the source material at - most OCR software (Acrobat included) require at least 300dpi resolution to accurately capture the text. Also, coloured backgrounds and odd fonts will sometimes confuzzle OCR software.

You could also Google for "OCR online free" and see if there is any similar sites like PDF Unlock or PDF Merge out there that do online OCR of PDFs.

I've actually not had a LOT of problems OCRing the god-awfully scanned journal articles that I've gotten from the UW library. They're all fairly low DPI and copies of copies of copies. Sometimes you get garbled words but for the most part it's fairly accurate.

If OCR was perfect, google wouldn't have bothered implementing that captcha-code system to translate words

  On 5/21/2012 at 6:14 PM, disparaissant said:

I've actually not had a LOT of problems OCRing the god-awfully scanned journal articles that I've gotten from the UW library. They're all fairly low DPI and copies of copies of copies. Sometimes you get garbled words but for the most part it's fairly accurate.

You're fortunate (or Adobe's improved the OCR since Acrobat 8, which is the version we use at work here), as I've always had issues with it catching some of the poorer-scanned text.

eh, it seems online OCR gizmos can only convert files <2MB. que pena