Gets the array of vision Y offsets, which represents alignment offsets in Y per column of the vision image.