python实现读取学术论文PDF文件内容

import PyPDF2

# 打开PDF文件
with open('ELLK-Net_An_Efficient_Lightweight_Large_Kernel_Network_for_SAR_Ship_Detection.pdf', 'rb') as file:
    reader = PyPDF2.PdfReader(file)
    # 获取页面数
    num_pages = len(reader.pages)
    print(f"页面数: {num_pages}")
    
    # 读取每一页的文本
    for i in range(num_pages):
        page = reader.pages[i]
        text = page.extract_text()
        print(f"第 {i + 1} 页的内容:\n{text}")

import pdfplumber

# 打开PDF文件
with pdfplumber.open('example.pdf') as pdf:
    for i in range(len(pdf.pages)):
        page = pdf.pages[i]
        text = page.extract_text()
        print(f"第 {i + 1} 页的内容:\n{text}")

import fitz  # PyMuPDF
from PIL import Image
import io
# 打开PDF文件
doc = fitz.open('ELLK-Net_An_Efficient_Lightweight_Large_Kernel_Network_for_SAR_Ship_Detection.pdf')
for page_number,page in enumerate(doc):
    text = page.get_text()
    print(f"第 {page.number + 1} 页的内容:\n{text}")
    # 获取页面中的所有图片
    image_list = page.get_images(full=True)
    # 遍历所有图片
    for img_index, img in enumerate(image_list):
        # 图片的XREF
        xref = img[0]
        # 获取图片对象
        base_image = doc.extract_image(xref)
        # 获取图片的二进制数据
        image_bytes = base_image["image"]
        # 将二进制数据转为PIL图像
        image = Image.open(io.BytesIO(image_bytes))
        # # 左右翻转
        # image = image.transpose(method=Image.FLIP_LEFT_RIGHT)
        # 上下翻转
        image = image.transpose(method=Image.FLIP_TOP_BOTTOM)
        # 获取图片的扩展名
        image_ext = base_image["ext"]
        # 保存图片
        # with open(f"page_{page_number + 1}_img_{img_index + 1}.{image_ext}", "wb") as image_file:
        #     image_file.write(image_bytes)
        image.save(f"page_{page_number + 1}_img_{img_index + 1}.{image_ext}")
        print(f"第 {page_number + 1} 页中提取的图片 {img_index + 1} 已保存。")

from pdfminer.high_level import extract_text

# 提取文本
text = extract_text('example.pdf')
print(text)