Python OCR

I want to convert jp2 file ( ex. newspapers) into text file. Now so far i have used pytesseract to do that and i was able to convert jp2 into text file with pytesseract but the issue is pytesseract not scanning jp2 file column by column. I want someone to help me reading jp2 file column by column. for this you can have your own approach or one of the approch in my mind is to split jp2 file by horizontal and vertical lines and then read each splitted file one by one.