Text extraction from PDF using 3 different methods

pdf-to-text.py

import pypdf
import easyocr
import pytesseract
import os
import fitz

pytesseract.pytesseract.tesseract_cmd = r"C:\Users\{USER_NAME}\AppData\Local\Programs\Tesseract-OCR\tesseract.exe"

FILE_PATH = "2024-V4.pdf"


def readPdf(pdf_path):
    try:
        content = pypdf.PdfReader(pdf_path)
        first_page = content.pages[0]
        print(first_page.extract_text())
    except:
        print("There was a problem in reading PDF file")

def ocrPdf(pdf_path, method):

    extracted_text = ""

    # OCR reader
    reader = easyocr.Reader(["en"], True)

    # Convert PDF to images
    pdf = fitz.open(pdf_path)
    for index, page in enumerate(pdf.pages()):
        image = page.get_pixmap()
        image.pil_save(f"ocr-image-test-{index}.png")

    # Read Text from images
    for file in os.listdir(os.getcwd()):
        if 'ocr-image-test-' in file:
            if method == "easyocr":
                content = reader.readtext(file, detail=False)
                extracted_text += f' {" ".join(content)}'
            
            if method == "pytesseract":
                content = pytesseract.image_to_string(file)
                extracted_text += content


    # Delete garbage
    for file in os.listdir(os.getcwd()):
        if 'ocr-image-test-' in file:
            os.remove(file)

    print(extracted_text)

def main(pdf_file):
    # Check if file exists
    exist = os.path.exists(pdf_file)
    if not exist:
        print("CV file does not exist")
        return
    
    # readPdf(pdf_file)
    # ocrPdf(pdf_file, "pytesseract")
    ocrPdf(pdf_file, "easyocr")


main(FILE_PATH)