如何在Python中将扫描的PDF转换为可搜索的PDF？[环境:Windows]

import os
from PIL import Image
import pytesseract
from PyPDF2 import PdfWriter

# 安装依赖库：pip install PyPDF2 pytesseract pillow

# 设置Tesseract的安装路径（根据实际安装路径进行设置）
pytesseract.pytesseract.tesseract_cmd = 'C:/Program Files/Tesseract-OCR/tesseract.exe'

# 扫描PDF文件路径
scan_pdf_path = 'path/to/scan.pdf'

# 临时图像文件夹路径
temp_image_folder = 'path/to/temp_images/'

# 输出PDF文件路径
output_pdf_path = 'path/to/output.pdf'

# 创建临时图像文件夹
os.makedirs(temp_image_folder, exist_ok=True)

# 将扫描的PDF转换为图像文件
def convert_pdf_to_images(pdf_path, image_folder):
    images = []
    with Image.open(pdf_path) as pdf:
        for i in range(pdf.n_frames):
            pdf.seek(i)
            image_path = os.path.join(image_folder, f'page_{i}.png')
            pdf.save(image_path, 'PNG')
            images.append(image_path)
    return images

# 图像文字识别
def ocr_images(images):
    text = ''
    for image in images:
        text += pytesseract.image_to_string(Image.open(image))
    return text

# 创建可搜索的PDF
def create_searchable_pdf(text, output_path):
    pdf_writer = PdfWriter()
    pdf_writer.add_blank_page()
    pdf_writer.add_page()
    pdf_writer.add_text(0, 0, text)
    with open(output_path, 'wb') as output_pdf:
        pdf_writer.write(output_pdf)

# 将扫描的PDF转换为图像文件
images = convert_pdf_to_images(scan_pdf_path, temp_image_folder)

# 图像文字识别
text = ocr_images(images)

# 创建可搜索的PDF
create_searchable_pdf(text, output_pdf_path)

# 清理临时图像文件夹
for image in images:
    os.remove(image)
os.rmdir(temp_image_folder)

print('转换完成！可搜索的PDF文件已保存到', output_pdf_path)