Python OpenCV 去除PDF（图片）中指定颜色水印_python识别pdf里指定颜色的内容

对比图。水印去除前后；   

 

                                                                     

 

 代码：

import cv2
import numpy as np
 
imgs = cv2.imread("1.png")
image = imgs.copy()
# images = imgs.copy()
# print(image.shape)
# rows, cols = image.shape[:2]
# print(rows, cols)
red_minus_blue = image[:, :, 2] - image[:, :, 0]
cv2.imwrite("res.jpg",image[:, :, 2])
 
 
# red_minus_green = image[:, :, 2] - image[:, :, 1]
 
 
# red_minus_blue = red_minus_blue >= 20
# red_minus_green = red_minus_green >= 20
 
# red = image[:, :, 2] >= np.mean(image[:, :, 2])/2
 
# mask = red_minus_green & red_minus_blue & red
# print(mask)
# images[mask,:]=255
# mask = (1-mask).astype(np.bool)
# print(mask)
# image[mask, :] = 255
 
# # stack = np.vstack([imgs,image,images])
# # cv2.imshow("stack", images)
# cv2.imwrite("result.jpg",images)
# # cv2.imshow("orgin", imgs)
# # cv2.imshow("red", image)
# cv2.imshow("delete_red", images)
cv2.waitKey()
 
# 红色像素值最大，且大于阈值（中值）
# 其他通道像素值的距离比较小，且与红色像素值的距离比率较大，且大于阈值（自己调）。

 

 

 

Python 获取pdf文件中的所有图片，保存到文件夹中。

pip install fitz
 
pip install pymupdf

 

import fitz
 
 
def func(doc):
    for i in range(len(doc)):
        imglist = doc.getPageImageList(i)
        for j, img in enumerate(imglist):
            xref = img[0]
            pix = fitz.Pixmap(doc, xref)   # make pixmap from image
            if pix.n - pix.alpha < 4:      # can be saved as PNG
                pix.writePNG("p%s-%s.png" % (i+1, j+1))
            else:                          # CMYK: must convert first
                pix0 = fitz.Pixmap(fitz.csRGB, pix)
                pix0.writePNG("p%s-%s.png" % (i+1, j+1))
                pix0 = None                # free Pixmap resources
            pix = None                     # free Pixmap resources
 
 
if __name__ == "__main__":
    func(doc=fitz.open('11.pdf'))         # input the path of pdf file
 
 
###参考博客 https://blog.csdn.net/qq_42951560/article/details/109609662

 

批量去除水印：

def remove():
    """
    去除水印
    """
    list_file=os.listdir('./pic')
    print(list_file)
    for i in list_file:
        imgs = cv2.imread('./pic/'+i)
        image = imgs.copy()
        # print(imgs)
        cv2.imwrite("./res/"+i,image[:, :, 2])

 

完整代码：

目录结构：

import fitz
import os
import cv2
import numpy as np
 
 
def func(doc):
    for i in range(len(doc)):
        imglist = doc.getPageImageList(i)
        for j, img in enumerate(imglist):
            xref = img[0]
            pix = fitz.Pixmap(doc, xref)   # make pixmap from image
            if pix.n - pix.alpha < 4:      # can be saved as PNG
                pix.writePNG("./pic/p%s-%s.png" % (i+1, j+1))
            else:                          # CMYK: must convert first
                pix0 = fitz.Pixmap(fitz.csRGB, pix)
                pix0.writePNG("./pic/p%s-%s.png" % (i+1, j+1))
                pix0 = None                # free Pixmap resources
            pix = None                     # free Pixmap resources
 
 
def remove():
    """
    去除水印
    """
    list_file=os.listdir('./pic')
    print(list_file)
    for i in list_file:
        imgs = cv2.imread('./pic/'+i)
        image = imgs.copy()
        # print(imgs)
        cv2.imwrite("./res/"+i,image[:, :, 2])
 
 
 
 
 
 
 
if __name__ == "__main__":
    # 第一步 将pdf中所有图片写入文件夹
    # func(doc=fitz.open('11.pdf'))         # input the path of pdf file
    # 第二步  去除文件夹内所有图片的红色水印
    # remove()
 
 

 

使用金山wps工具就可以将文件夹内的图片合成pdf文档了。

 

参考博客：https://blog.csdn.net/a362682954/article/details/105628198/

参考博客 ：https://blog.csdn.net/qq_42951560/article/details/109609662