I have some images that have up and down arrows beside some text, and was wondering how I can utilize/optimize tesseract-ocr to detect these symbols. The current output classifies this arrow as a 'T' and some other cases were classified as ' | '. I'm hoping it would return ↑ sweating.
*Update: All up arrows are classified as T it seems, but down arrows are classified differently. (Picture #2)
>>> print(pytesseract.image_to_string(image))
T sweating
Additionally, down arrows get classified differently.
,
>>> print(pytesseract.image_to_string(image))
4 heart rate
And other examples sometimes recognize the down arrow as L or |.
I understand there are various additional languages I can use in the OCR, so my question is are there any languages/configurations I can add to detect arrows?