[bcab] Re: FineReader OCR

Tony, for what it's worth I find it can simplify things to convert scanned 
images to grey-scale and then adjust the brightness and shadow settings in a 
graphics program so that the text stands out more - and sometimes it makes it 
clearer if you make the image negative. Depends so much on all your hardware 
and your vision, I suppose.

Someone shoed me the 'document' mode on their digital camera, and I must say 
there's a lot of potential there for creating the equivalent of a scanned 
image. I think the spec of the camera was about 6 megapixels, which is fairly 
standard now.

Vince.
  ----- Original Message ----- 
  From: Tony M Dart 
  To: bcab@xxxxxxxxxxxxx 
  Sent: Wednesday, February 28, 2007 2:31 PM
  Subject: [bcab] FineReader OCR


  Hi

  I have recently been doing some tests with FineReader OCR on difficult to 
read material - text over coloured backgrounds and photographs. (Specifically 
the backs of paperback books).

  I have been using both FineReader 7 and FineReader 8. Oddly enough, 7 
performs this much better than 8 - it is faster and far, far more accurate 
(often 100%!).

  I have also found that successive scans on my cheapo Xerox flatbed scanner 
give different errors, so I assume this is down to fluctuations in the 
scanner's physical performance. Successive photos with a digital camera do not 
give this problem.

  I would be deeply interested to hear of anyone elses experiences, experiments 
and setups for OCR.

  Regards
  Tony


I'm protected by SpamBrave
http://www.spambrave.com/

Other related posts: