|
| 1 | +#!/usr/bin/env python3 |
| 2 | + |
| 3 | +importfitz#pip install pymupdf |
| 4 | +importre |
| 5 | +importos |
| 6 | + |
| 7 | + |
| 8 | +deffind_imag(path,img_path): |
| 9 | + |
| 10 | +checkXO=r"/Type(?= */XObject)" |
| 11 | +checkIM=r"/Subtype(?= */Image)" |
| 12 | + |
| 13 | +pdf=fitz.open(path) |
| 14 | + |
| 15 | +img_count=0 |
| 16 | +len_XREF=pdf._getXrefLength() |
| 17 | + |
| 18 | +print("文件名:{}, 页数: {}, 对象: {}".format(path,len(pdf),len_XREF-1)) |
| 19 | + |
| 20 | +foriinrange(1,len_XREF): |
| 21 | +text=pdf._getXrefString(i) |
| 22 | +isXObject=re.search(checkXO,text) |
| 23 | + |
| 24 | +# 使用正则表达式查看是否是图片 |
| 25 | +isImage=re.search(checkIM,text) |
| 26 | + |
| 27 | +# 如果不是对象也不是图片，则continue |
| 28 | +ifnotisXObjectornotisImage: |
| 29 | +continue |
| 30 | +img_count+=1 |
| 31 | +# 根据索引生成图像 |
| 32 | +pix=fitz.Pixmap(pdf,i) |
| 33 | + |
| 34 | +new_name=path.replace('\\','_')+"_img{}.png".format(img_count) |
| 35 | +new_name=new_name.replace(':','') |
| 36 | + |
| 37 | +# 如果pix.n<5,可以直接存为PNG |
| 38 | +ifpix.n<5: |
| 39 | +pix.writePNG(os.path.join(img_path,new_name)) |
| 40 | + |
| 41 | +else: |
| 42 | +pix0=fitz.Pixmap(fitz.csRGB,pix) |
| 43 | +pix0.writePNG(os.path.join(img_path,new_name)) |
| 44 | +pix0=None |
| 45 | + |
| 46 | +pix=None |
| 47 | + |
| 48 | +print("提取了{}张图片".format(img_count)) |
| 49 | + |
| 50 | + |
| 51 | +if__name__=='__main__': |
| 52 | +pdf_path=r'test.pdf' |
| 53 | +img_path=r'img' |
| 54 | +m=find_imag(pdf_path,img_path) |