How to Read Contents of PDF using OCR in Python

To read the contents of a PDF using Optical Character Recognition (OCR) in Python, you can use the Tesseract OCR library, which is a popular and open-source OCR engine. Here are the steps to perform OCR on a PDF document:

1. Install Tesseract OCR: Before you can use Tesseract OCR in Python, you need to install Tesseract on your system. You can download Tesseract from the official website (https://github.com/tesseract-ocr/tesseract) and follow the installation instructions for your operating system. Additionally, you’ll need to install the pytesseract library, which provides a Python interface to Tesseract:

   pip install pytesseract

2. Install PDF Libraries: If your PDF contains scanned images (i.e., not selectable text), you’ll also need a library to extract images from the PDF. One popular choice is PyMuPDF (also known as Fitz). You can install it with:

   pip install PyMuPDF

3. Perform OCR on the PDF: Here’s a Python script that uses PyMuPDF to extract images from a PDF and then uses pytesseract to perform OCR on each image:

   import fitz  # PyMuPDF
   import pytesseract

   # Open the PDF file
   pdf_file = "your_pdf_file.pdf"
   pdf_document = fitz.open(pdf_file)

   # Initialize an empty string to store the extracted text
   extracted_text = ""

   # Iterate through each page in the PDF
   for page_num in range(pdf_document.page_count):
       page = pdf_document.load_page(page_num)
       image_list = page.get_images(full=True)

       # Extract text from each image on the page
       for img_index, img in enumerate(image_list):
           xref = img[0]
           base_image = pdf_document.extract_image(xref)
           image_data = base_image["image"]

           # Perform OCR using pytesseract on the image
           ocr_text = pytesseract.image_to_string(image_data, lang="eng")  # Specify language as needed

           # Append the extracted text to the result
           extracted_text += ocr_text + "\n"

   # Close the PDF file
   pdf_document.close()

   # Print or save the extracted text
   print(extracted_text)

Make sure to replace "your_pdf_file.pdf" with the path to your PDF file. This script will extract text from images within the PDF using OCR and store it in the extracted_text variable.

4. Additional Configuration: You can configure Tesseract OCR for different languages and additional settings. The lang parameter in pytesseract.image_to_string() specifies the language. You may need to download language data files for the languages you intend to use.

   sudo apt-get install tesseract-ocr-{language_code}

Replace {language_code} with the appropriate language code, such as eng for English.

5. Handling OCR Results: Once you have extracted the text using OCR, you can process, analyze, or save it as needed.

Keep in mind that OCR accuracy may vary depending on the quality of the scanned images and the clarity of the text. It’s important to review and possibly correct the extracted text, especially for complex or critical documents.