Python读取PDF中的图片：完整教程与代码示例 | Python PDF处理指南

http://lanqiu.cqfowei.com 
http://zhibo.cqfowei.com 
http://jrs.cqfowei.com 
http://tv.cqfowei.com 
http://qq.cqfowei.com 

pip install PyPDF2

pip install pdf2image

import PyPDF2
from PIL import Image
import os

def extract_images_with_pypdf2(pdf_path, output_folder):
    # 确保输出目录存在
    os.makedirs(output_folder, exist_ok=True)
    
    # 打开PDF文件
    pdf_file = open(pdf_path, 'rb')
    pdf_reader = PyPDF2.PdfReader(pdf_file)
    
    # 遍历所有页面
    for page_num in range(len(pdf_reader.pages)):
        page = pdf_reader.pages[page_num]
        
        # 获取页面中的XObject
        if '/XObject' in page['/Resources']:
            x_object = page['/Resources']['/XObject'].get_object()
            
            for obj in x_object:
                if x_object[obj]['/Subtype'] == '/Image':
                    # 获取图像对象
                    img = x_object[obj]
                    
                    # 获取图像数据
                    img_data = img.get_data()
                    
                    # 根据图像格式保存
                    if '/Filter' in img:
                        if img['/Filter'] == '/FlateDecode':
                            # 保存为PNG
                            img_path = os.path.join(output_folder, 
                                                  f"page_{page_num+1}_img_{obj[1:]}.png")
                            with open(img_path, 'wb') as img_file:
                                img_file.write(img_data)
                        elif img['/Filter'] == '/DCTDecode':
                            # 保存为JPEG
                            img_path = os.path.join(output_folder, 
                                                  f"page_{page_num+1}_img_{obj[1:]}.jpg")
                            with open(img_path, 'wb') as img_file:
                                img_file.write(img_data)
                        elif img['/Filter'] == '/JPXDecode':
                            # 保存为JPEG 2000
                            img_path = os.path.join(output_folder, 
                                                  f"page_{page_num+1}_img_{obj[1:]}.jp2")
                            with open(img_path, 'wb') as img_file:
                                img_file.write(img_data)
                    else:
                        # 尝试保存为PNG
                        img_path = os.path.join(output_folder, 
                                              f"page_{page_num+1}_img_{obj[1:]}.png")
                        with open(img_path, 'wb') as img_file:
                            img_file.write(img_data)
    
    pdf_file.close()
    print(f"图片已提取到: {output_folder}")

# 使用示例
extract_images_with_pypdf2('example.pdf', 'extracted_images')

from pdf2image import convert_from_path
import os

def extract_images_with_pdf2image(pdf_path, output_folder, dpi=200, 
                                 fmt='jpeg', thread_count=4):
    """
    使用pdf2image从PDF中提取图片
    
    参数:
        pdf_path: PDF文件路径
        output_folder: 输出目录
        dpi: 图像分辨率(默认为200)
        fmt: 输出格式('jpeg', 'png', 'tiff')
        thread_count: 使用的线程数
    """
    # 确保输出目录存在
    os.makedirs(output_folder, exist_ok=True)
    
    # 转换PDF为图像
    images = convert_from_path(pdf_path, 
                              dpi=dpi, 
                              fmt=fmt,
                              output_folder=output_folder,
                              output_file='page',
                              paths_only=True,
                              thread_count=thread_count)
    
    print(f"成功提取 {len(images)} 张图片到 {output_folder}")

# 使用示例
extract_images_with_pdf2image('example.pdf', 'pdf_images', dpi=300)

images = convert_from_path('doc.pdf', dpi=300)

# 每次只处理一页
for page_num in range(10):  # 处理前10页
    images = convert_from_path('large.pdf', 
                             first_page=page_num+1, 
                             last_page=page_num+1)

images = convert_from_path('doc.pdf', 
                         poppler_path=r'C:\path\to\poppler\bin')