text detection before full ocr

[Stefano]

Hello,

I started to play with this great software as I would like to add OCR functionality to my service. I am concerned about the time needed to perform a full recognition on an image and I was wondering if I should first just quickly check if probability there is text is high enough; if so, I would do a full OCR on it. However, maybe this would be overkilled because tesseract does it already in the pre-processing phase. Is this the case? is there any parameter to tweak to optimize the process so that it exists as soon as possible if no text is detected?

I searched similar question but could not find an answer. Sorry if I missed the information if already available.

Thanks in advance,
Stefano