Computer vision script in python

Build a function in python that takes an image as an input and returns a modified picture using EasyOCR/OpenCV/SimpleCV library (final tool or set of tools is your choice but it has to be open source).
Inputs:
- a picture containing text (also non-Latin alphabet should be supported; spanish / greek / cyrillic ), a set of empty boxes located in one column, and a hand written mark (a check mark or a cross or different)
- text string
- code for the alphabet / language
Outputs:
- modified picture
- text string
Modification:
if the input picture contains the input text string:
if the handwritten mark is located in the box closest to the input string:
do nothing, return original image
otherwise:
locate an empty box closest to the input string
move the handwritten mark from the original box to the located box
the original box should be clear
return text string corresponding to the originally marked box and the modified picture
otherwise:
return an error

Should work for arbitrary picture and values of the input text string. The picture will be a photo so there might be some slope.

Contact me in DM and share the progress on the work every 3 days
Provide comments for key lines in the code