0

I'm having problems with tesseract conducting OCR of numbers. This picture

enter image description here

is recognised as

‘I .38)

I'm using -psm 6 as parameter. Are there any better ways to recognise numbers? Do I need to configure tesseract or is there any additional pre processing needed of the image?

Nickpick
  • 6,163
  • 16
  • 65
  • 116

1 Answers1

3

I rescaled to 300DPI and got the correct result.

nguyenq
  • 8,212
  • 1
  • 16
  • 16
  • Is rescaling generally something that is recommended? – Nickpick Feb 21 '16 at 09:57
  • It's normally required for screenshots for their low resolutions. https://github.com/tesseract-ocr/tesseract/wiki/FAQ#is-there-a-minimum-text-size-it-wont-read-screen-text – nguyenq Feb 21 '16 at 15:01