Tolerance when recognizing spaces

[Lemvigh]

Is there a parameter controlling the with of a space character? Our
biggest problem regarding the OCR is concatenated words. There is a
lot of spaces that Tesseract doesn't find. Is there a way the
recognition of spaces can be tuned or a variable controlling the
minimum with of a space character?

Regards,
Lemvigh

[Ray Smith]

Yes, but I don't know what will affect it the way you want in your particular circumstances.

I suggest looking at textord/tospace.cpp, where you will find lots of control parameters near the top. You could try adjusting some of these...

Ray.