使用Python提高PDF分辨率的完整指南

pip install PyMuPDF Pillow

import fitz  # PyMuPDF的别名

# 加载PDF文件
def load_pdf(file_path):
    pdf_document = fitz.open(file_path)
    return pdf_document

# 示例：加载名为sample.pdf的PDF文件
pdf_file = load_pdf("sample.pdf")

from PIL import Image

def increase_resolution(pdf_document, scale=2):
    for page_num in range(len(pdf_document)):
        # 获取当前页
        page = pdf_document[page_num]
        
        # 获取页的矩形大小
        rect = page.rect
        # 裁剪出指定区域的图像
        pix = page.get_pixmap(matrix=fitz.Matrix(scale, scale))  
        
        # 将pix转换为图像
        img = Image.frombytes("RGB", [pix.width, pix.height], pix.samples)
        # 可以根据需要处理图像
        
        # 保存或处理图像
        img.save(f"page_{page_num + 1}.png")  # 保存为PNG格式的图像

# 示例：提高分辨率
increase_resolution(pdf_file, scale=2)

def save_pdf(pdf_document, output_path):
    pdf_document.save(output_path)

# 示例：保存新的PDF文件
save_pdf(pdf_file, "output.pdf")

pdf_file.close()  # 关闭PDF文件

erDiagram
    PDFDocument {
        +string filePath
        +string title
    }
    
    Image {
        +string filePath
        +string format
    }
    
    PDFDocument ||--o{ Image : contains

import fitz  # PyMuPDF的别名
from PIL import Image

def load_pdf(file_path):
    pdf_document = fitz.open(file_path)
    return pdf_document

def increase_resolution(pdf_document, scale=2):
    for page_num in range(len(pdf_document)):
        page = pdf_document[page_num]
        rect = page.rect
        pix = page.get_pixmap(matrix=fitz.Matrix(scale, scale))
        img = Image.frombytes("RGB", [pix.width, pix.height], pix.samples)
        img.save(f"page_{page_num + 1}.png")

def save_pdf(pdf_document, output_path):
    pdf_document.save(output_path)

# 主程序
if __name__ == "__main__":
    pdf_file = load_pdf("sample.pdf")
    increase_resolution(pdf_file, scale=2)
    save_pdf(pdf_file, "output.pdf")
    pdf_file.close()

import fitz
import numpy as np
from PIL import Image
import cv2
from tqdm import tqdm
import io
 
pdf_doc = fitz.open('/Users/linql/Desktop/11.pdf')
output_doc = fitz.open()
for orig_page in tqdm(pdf_doc):
    zoom = 2
    mat = fitz.Matrix(zoom, zoom)
    pix = orig_page.get_pixmap(matrix=mat)
    img = Image.frombytes("RGB",
                          [pix.width, pix.height],
                          pix.samples)
    img = img.convert('L')
    cleaned_page_array = cv2.adaptiveThreshold(np.array(img),
                                               255,
                                               # cv2.ADAPTIVE_THRESH_MEAN_C, #基于邻域均值的自适应阈值。
                                               cv2.ADAPTIVE_THRESH_GAUSSIAN_C,#基于邻域加权平均的自适应阈值。
                                               cv2.THRESH_BINARY,
                                               19,
                                               15)
    new_image = Image.fromarray(cleaned_page_array)
    fp = io.BytesIO()
    new_image.save(fp, "JPEG")
    page = output_doc.new_page(width=new_image.size[0], height=new_image.size[1])
    page.insert_image(page.rect, stream=fp.getvalue())
 
output_doc.save("output_file.pdf")