1

I believe the text result of OCR not only depends on how good the OCR library is but also depends on the image to scan.

My question is that, for an image, how could I make it easier to recognize?

Change the original color? (for example, to grey)

Change the scale/size? (for example, bigger)

Change the area to recognize? (for example, set the distance to top/left side)

Any other idea?

Jack He
  • 1,683
  • 3
  • 18
  • 29
  • 1
    Probably don't bother doing anything that doesn't add information. The OCR library is capable of converting a source image to greyscale (for example) before processing. So if that consistently helps, it's probably doing it internally already. If you can crop the image intelligently, or perform any other transformation using information that OCR library doesn't have, that might help. – Blorgbeard May 13 '16 at 01:31
  • @Blorgbeard Thanks for your response. I am using Tesseract. I am not quite sure if this library has processed the thing you mentioned. Any information about it? – Jack He May 13 '16 at 01:45
  • 2
    Nope, have you read the docs? 15 seconds of googling reveals what looks like a good start: http://stackoverflow.com/questions/9480013/image-processing-to-improve-tesseract-ocr-accuracy – Blorgbeard May 13 '16 at 01:52
  • @Blorgbeard Hey, man, that's really what I want. Thanks a lot! – Jack He May 13 '16 at 04:18

0 Answers0