How can I tell Google Document AI Enterprise OCR to always assume one column?
My text (scans of old books) are always one column. However, due to layout, (lots of) whitespace, and inline figures, Google Document AI Enterprise OCR sometimes splits the text into two columns. This causes words to be jumbled out of order:
P: The quick brown fox
jumped over the lazy hen.
becomes
P1: The quick brown jumped over the
P2: fox lazy hen.
Since my text is always one column (though the left and right bounds move, due to whitespace, formatting, and inline images), is there a way to tell this to Google Document AI Enterprise OCR? Or a way to fix it post-OCR?