利用python將圖片版PDF轉(zhuǎn)文字版PDF

更新時(shí)間：2019年05月03日 15:59:20 投稿：WDC

今天為大家介紹一下如何使用利用python將圖片版PDF轉(zhuǎn)文字版PDF，這里我們需要用到python3.6,pypdf2,ghostscript,PythonMagick,百度文字識(shí)別服務(wù)和pdfkit

圖片版PDF無法復(fù)制,轉(zhuǎn)化成文字版的PDF后使用更方便.

我們需要用到python3.6,pypdf2,ghostscript,PythonMagick,百度文字識(shí)別服務(wù)和pdfkit.

安裝

安裝python3.6 略

安裝ghostscript

https://ghostscript.com/download/gsdnld.html

安裝wkhtmltopdf

https://wkhtmltopdf.org/downloads.html

pip安裝PyPDF2,ghostscript,baidu-aip,pdfkit

pip install PyPDF2
pip install ghostscript
pip install baidu-aip
pip install pdfkit

pip安裝PythonMagick

https://www.lfd.uci.edu/~gohlke/pythonlibs/

cd 下載目錄
pip install PythonMagick‑0.9.13‑cp36‑cp36m‑win_amd64.whl

pypdf2用于拆分和合并PDF

示例代碼如下:

#導(dǎo)入PdfFileReader和PdfFileWriter
from PyPDF2 import PdfFileReader, PdfFileWriter
#獲取一個(gè)pdf對(duì)象
pdf_input = PdfFileReader(open(r'pdf路徑', 'rb'))
#獲取pdf頁數(shù)
page_count = pdf_input.getNumPages()
#獲取pdf第四頁的內(nèi)容
page = pdf_input.getPage(3)
page['/Contents']
#獲取一個(gè)pdfWriter對(duì)象
pdf_output = PdfFileWriter()
# 將一個(gè) PageObject 加入到 PdfFileWriter 中
pdf_output.addPage(page)
#把新pdf保存
pdf_output.write(open(r'新pdf路徑','wb'))

PythonMagick用于將單頁P(yáng)DF轉(zhuǎn)化為jpg

百度云-文字識(shí)別-python SDK

每天有500次免費(fèi)的識(shí)別
示例代碼如下:

#導(dǎo)入baidu-aip
from aip import AipOcr
#https://console.bce.baidu.com/#/index/overview
#產(chǎn)品服務(wù)->人工智能->文字識(shí)別->創(chuàng)建應(yīng)用
#獲取以下三個(gè)值
APP_ID = '??'
API_KEY = '??'
SECRET_KEY = '?? '
#新建一個(gè)AipOcr
client = AipOcr(APP_ID, API_KEY, SECRET_KEY)
#讀取本地圖片的函數(shù)
def get_file_content(filePath):
  with open(filePath, 'rb') as fp:
    return fp.read()
#讀取本地圖片
image = get_file_content('p1.jpg')
#可選參數(shù)
options = {}
options["language_type"] = "CHN_ENG"
options["detect_direction"] = "true"
options["detect_language"] = "true"
options["probability"] = "true"
#通用文字識(shí)別
client.basicGeneral(image, options)

#讀取網(wǎng)絡(luò)圖片
url = "https://note.youdao.com/yws/public/resource/1577071c1ffa2b6bf4e238ef6dbcfbf5/xmlnote/E5A19BEDFEBA4879B217C5BBF53B0245/22138"
#可選參數(shù)
options = {}
options["language_type"] = "CHN_ENG"
options["detect_direction"] = "true"
options["detect_language"] = "true"
options["probability"] = "true"
#通用文字識(shí)別
client.basicGeneralUrl(url, options)

#讀取本地表格圖片的函數(shù)
def get_file_content(filePath):
  with open(filePath, 'rb') as fp:
    return fp.read()
#讀取本地表格圖片
image = get_file_content('p2.jpg')
#可選參數(shù)
options = {}
options["language_type"] = "CHN_ENG"
options["detect_direction"] = "true"
options["detect_language"] = "true"
options["probability"] = "true"
#通用文字識(shí)別
client.basicGeneral(image, options)
#讀取表格分割效果較差!

pdfkit用于利用字符串生成pdf
示例代碼如下:

#pdfkit安裝位置設(shè)置
path_wk = r'pdfkit安裝位置設(shè)置'
pdfkit_config = pdfkit.configuration(wkhtmltopdf = path_wk)
#pdfkit參數(shù)
pdfkit_options = {'encoding': 'UTF-8',}
#制作PDF
pdfkit.from_string(('string'),'D:\test.pdf',configuration=pdfkit_config,options=pdfkit_options)

完整代碼如下

#導(dǎo)入所需包

#os,操作文件和路徑
import os
#ghostscript,代碼簡(jiǎn)化
import ghostscript
#pypdf2,拆分pdf
from PyPDF2 import PdfFileReader, PdfFileWriter
#PythonMagick,單頁P(yáng)DF轉(zhuǎn)圖片
from PythonMagick import Image
#baidu-aip,百度文字識(shí)別
from aip import AipOcr
#pdfkit,字符串制作PDF
import pdfkit

#參數(shù)

path='??'
pdfname='??'
DPI='85'
#https://console.bce.baidu.com/#/index/overview
#產(chǎn)品服務(wù)->人工智能->文字識(shí)別->創(chuàng)建應(yīng)用
#獲取以下三個(gè)值
APP_ID = '??'
API_KEY = '??'
SECRET_KEY = '?? '
#pdfkit安裝位置設(shè)置
path_wk = r'pdfkit安裝位置設(shè)置'
pdfkit_config = pdfkit.configuration(wkhtmltopdf = path_wk)
#pdfkit參數(shù)
pdfkit_options = {'encoding': 'UTF-8',}

#PDF轉(zhuǎn)化為圖片

os.chdir(path)
pdf_input = PdfFileReader(open(pdfname, 'rb'))
#自動(dòng)獲取PDF頁數(shù)
page_count = pdf_input.getNumPages()
page_range=range(page_count)
#也可以手工指定PDF需要轉(zhuǎn)換的頁數(shù)
#page_range=range(0,100)
#使用PyPDF和ghostscript
#==超級(jí)好用,超級(jí)直觀,超級(jí)短==
for page_num in page_range:
  im = Image()
  im.density(DPI)
  im.read(pdfname + '[' + str(page_num) +']')
  im.write(str(page_num)+ '.jpg')

#圖片轉(zhuǎn)化為字符串

#新建一個(gè)AipOcr
client = AipOcr(APP_ID, API_KEY, SECRET_KEY)
#讀取本地圖片的函數(shù)
def get_file_content(filePath):
  with open(filePath, 'rb') as fp:
    return fp.read()
#可選參數(shù)
options = {}
options["language_type"] = "CHN_ENG"
options["detect_direction"] = "false"
options["detect_language"] = "false"
options["probability"] = "false"
allteststr=[]
for page_num in page_range:
  #讀取本地圖片
  image = get_file_content(r'%s\%s.jpg' % (path,page_num))
  #通用文字識(shí)別,得到的是一個(gè)dict
  testjson=client.basicGeneral(image, options)
  teststr=''
  for x in testjson['words_result']:
    teststr=teststr+x['words']+'</br>'
  allteststr.append(teststr)

#字符串寫入PDF
for page_num in page_range:
  pdfkit.from_string((allteststr[page_num]),'%s.pdf' % (str(page_num)),configuration=pdfkit_config,options=pdfkit_options)
#合并單頁P(yáng)DF
pdf_output = PdfFileWriter()
for page_num in page_range:
  os.chdir(path)
  pdf_input = PdfFileReader(open('%s.pdf' % (str(page_num)), 'rb'))
  page = pdf_input.getPage(0)
  pdf_output.addPage(page)
pdf_output.write(open('newpdf.pdf','wb'))

以上就是為大家介紹的如何使用python3.6,pypdf2,ghostscript,PythonMagick,百度文字識(shí)別服務(wù)和pdfkit

您可能感興趣的文章: