KiwanOCR

This package takes a single PDF file or a list of PDF files and returns their content as a text file.

• Requirements
• Methods

Requirements

pip install or brew install

Make sure you have installed these dependencies:

• brew install tesseract
• brew install poppler
• pip pdf2images

import

Import the following:

• from PIL import Image
• import pytesseract ## python interface for tesseract
• import os ## navitage, create directories
• import shutil ## to delete the image folders with their imgs
• from pdf2image import convert_from_path ## to turn pdf to image
• import glob ## to glob files into a list
• from pathlib import Path ## to specify path to your files
• from natsort import natsorted, ns ## natural sorting
• import re ## for regex

Methods

Setup

1. pip install kiwanocr.
2. from kiwano import ocr

OCR a single PDF

ocr.ocr_file(file_name, output_file_name, language, resolution)

Arguments

• file_name: as a string
• output_file_name: as a string
• language: default is English (use tesseract --list-langs to retrieve langague codes )
• resolution: default is 300 dpi (use integer value between 100 and 1200)

OCR a list of PDFs

ocr.ocr_files(list_name, output_file_name, language, resolution)

• list_name: The only difference is to enter a list name

Output

A .txt file is placed in a output folder.