Dear researchers,
when I use tesseract for character recognition, there exist so many wrong results which the character width or height is about 4 pixels.
Is there any way to restrict the tesseract recognition by a range of character size?
thank you very much for your time.
Best Regards
Hi,
Do you use it alone? I use it with Xsane (with a script xsane2tess)
-I prescan
-I select with the mouse the area I want
-I scan and get OCR!
and I can directly find out the resolution by some settings in xsane... If to high or too low, tesseract bugs!
Dear Benjamin,
Thank you very much for your answer.
We have hundreds of thousand aerial photos, which we hope to read some serial numbers from them.
I am working in R.
when tesseract recognizes them some noises which are about 3 pixels width or height are reported wrongly.
I was thinking of how to tell the tesseract not select numbers which are out of a range size.
I searched for long time but could not find a solution.
I would be so appreciate if having any hint.
Best regards and thank you for your time
- Badges
- Science topic