Mobile-OCR-position-fix

Google's mobile OCR's output maybe jumbled when there are many lines of text in the image, this project aims at fixing it.

input should be a file in CSV format

br_left = bounding rectangle's left
br_top = bounding rectangle's top
br_right = bounding rectangle's right
br_bottom = bounding rectangle's bottom

<recognized_word1>,<br_left>,<br_top>,<br_right>,<br_bottom>
<recognized_word2>,<br_left>,<br_top>,<br_right>,<br_bottom>
<recognized_word3>,<br_left>,<br_top>,<br_right>,<br_bottom>

Running the program

python3 bounding.py <input_file>