使用PyPDF从PDF中提取图像，xObject中不带“/Filter”标记

Question

目前正在使用类似的方法从PDF中提取图像：

import PyPDF4
from PIL import Image
from pathlib import Path
import os

PDFFilePath = Path("somefile.pdf")
OutputFolder = "somedirectory"
pdfpage = 0

with open(PDFFilePath,'rb') as pdf_reader:
    pdf_object = PyPDF4.PdfFileReader(pdf_reader)
    PageFolder = Path(OutputFolder).joinpath(Path(PDFFilePath.stem + '.'+ str(pdfPage)))
    if not PageFolder.exists():
        os.makedirs(PageFolder)

    CurrentPage = pdf_object.getPage(pdfPage)
    xObject = CurrentPage['/Resources']['/XObject'].getObject()

    for obj_index,obj in enumerate(xObject):
        if xObject[obj]['/Subtype'] == '/Image':
            size = (xObject[obj]['/Width'], xObject[obj]['/Height'])
            data = xObject[obj].getData()
            if xObject[obj]['/ColorSpace'] == '/DeviceRGB':
                mode = "RGB"
            else:
                mode = "P"

            if xObject[obj]['/Filter'] == '/FlateDecode':
                img = Image.frombytes(mode, size, data)
                img.save(PageFolder.joinpath(Path(PDFFilePath).stem +"."+ str(pdfPage) + "."+ str(obj_index) + ".png"),'wb')
            elif xObject[obj]['/Filter'] == '/DCTDecode':
                img = open(PageFolder.joinpath(Path(PDFFilePath).stem +"."+ str(pdfPage) + "."+ str(obj_index)+ ".jpg"),'wb')
                img.write(data)
                img.close()
            elif xObject[obj]['/Filter'] == '/JPXDecode':
                img = open(PageFolder.joinpath(Path(PDFFilePath).stem +"."+ str(pdfPage) + "."+ str(obj_index)+ ".jp2"),'wb')
                img.write(data)
                img.close()
            elif xObject[obj]['/Filter'] == '/CCITTFaxDecode':
                img = open(PageFolder.joinpath(Path(PDFFilePath).stem +"."+ str(pdfPage) + "."+ str(obj_index)+ ".tiff"),'wb')
                img.write(data)
                img.close()

我在xObject[obj]['/Filter']部分遇到了一堆没有“/Filter”的PDF。我试图通过枕头从data = xObject[obj].getdata()中提取原始图像，但抛出了一个错误，它“没有足够的数据”。如果使用cv2.imdecode，OpenCV将返回None

提供的PDF是保密的，因此我无法提供样本

一个仍然使用PyPDF4的解决方案会很好

编辑：OpenCV图像阅读器

OpenCV部分（我从代码中删除了它，如果未检测到“/Filter”，它将被删除）

cv_color_space = cv2.IMREAD_COLOR if mode == "RGB" else cv2.IMREAD_GRAYSCALE
buf = np.frombuffer(data,np.uint8)
img = cv2.imdecode(buf,cv_color_space)
cv2.imwrite("outputfile.png",img)

Answer 1

图像显然是.tiff图像，但没有标题。 我找到这个：https://stackoverflow.com/a/34555343/13919892

我在代码中添加了此函数：

import struct

def tiff_header_for_CCITT(width, height, img_size, CCITT_group=4):
    tiff_header_struct = '<' + '2s' + 'h' + 'l' + 'h' + 'hhll' * 8 + 'h'
    return struct.pack(tiff_header_struct,
                       b'II',  # Byte order indication: Little indian
                       42,  # Version number (always 42)
                       8,  # Offset to first IFD
                       8,  # Number of tags in IFD
                       256, 4, 1, width,  # ImageWidth, LONG, 1, width
                       257, 4, 1, height,  # ImageLength, LONG, 1, lenght
                       258, 3, 1, 1,  # BitsPerSample, SHORT, 1, 1
                       259, 3, 1, CCITT_group,  # Compression, SHORT, 1, 4 = CCITT Group 4 fax encoding
                       262, 3, 1, 0,  # Threshholding, SHORT, 1, 0 = WhiteIsZero
                       273, 4, 1, struct.calcsize(tiff_header_struct),  # StripOffsets, LONG, 1, len of header
                       278, 4, 1, height,  # RowsPerStrip, LONG, 1, lenght
                       279, 4, 1, img_size,  # StripByteCounts, LONG, 1, size of image
                       0  # last IFD
                       )

然后将其添加到我的代码中：

if not '/Filter' in xObject[obj]:
    tiff_header = tiff_header_for_CCITT(size[0],size[1],len(data),1) # Using the group "1" because it works for some reason
    inv_data = bytes((~bit + 256 for bit in data)) # for some reason the bits are inverted?
    tiff_data = tiff_header + inv_data # Add the header to the inverted data
    # Write the tiff file
    img = open(PageFolder.joinpath(Path(PDFFilePath).stem +"."+ str(pdfPage) + "."+ str(obj_index)+ ".tiff"),'wb')
    img.write(tiff_data)
    img.close()
    continue

我需要知道如何识别位是否需要反转，或者使用什么“CCITT组”

我将把这个标记为答案，也许只是为这个打开一个新的问题