Explicitly set the font to be used for recognition by Tesseract-OCR

Question

I have documents which use only one font throughout the document. Different documents might have different fonts, but I know which document uses which font.

Is there an option to explicitly tell Tesseract-OCR which font to use during recognition for a given image?

To help debugging this, you can set `tessedit_debug_fonts` to 1 in order to determine how & what font tesseract identify — Lucas Cimon, Nov 25 '19 at 22:48

score 6 · Accepted Answer · edited Jul 04 '18 at 17:23

6

No, I don't think Tesseract supports such an option. What you can do is to train for one specific font and then specify that traineddata during recognition of your documents.

edited Jul 04 '18 at 17:23

Delimitry

2,987
4
30
39

answered Nov 02 '12 at 00:12

nguyenq

8,212
1
16
16

Explicitly set the font to be used for recognition by Tesseract-OCR

1 Answers1

Linked