环境版本: WIN10 | Python 3.6 | ImageMagick-6.9.9-38-Q8-x64-dll | Ghostscript 9.22 for Windows
整体思路:1.将PDF转为图片后进行文字识别 | 2.使用pdfminer解析pdf文件(准确率更高)
目录
在 github.com/UB-Mannheim/tesseract/wiki下载tesseract-xxx.exe文件后安装即可。需要注意的是,在选择安装组件时点开“Language data” 选上你要识别的语言,不选的话只能识别英文哟。
pip install pyocr
pip install Wand
pip install Pillow
若python版本为2.x需在http://pythonware.com/products/pil/下载Python Imaging Library (PIL) exe文件安装。
Wand依赖ImageMagick,ImageMagick依赖Ghostscript,去下面链接下载安装即可。
ImageMagick:http://docs.wand-py.org/en/latest/guide/install.html#install-imagemagick-on-windows
注意不要下载 ImageMagick 7.x版本,wand不支持。
Ghostscript :https://ghostscript.com/download/gsdnld.html
# CHANGE THIS IF TESSERACT IS NOT IN YOUR PATH, OR IS NAMED DIFFERENTLY
#TESSERACT_CMD = 'tesseract.exe' if os.name == 'nt' else 'tesseract'
#TESSERACT_CMD = 'D:/Program Files (x86)/Tesseract-OCR/tesseract.exe' if os.name == 'nt' else 'tesseract'
TESSERACT_CMD = os.environ["TESSDATA_PREFIX"] + '/tesseract.exe' if os.name == 'nt' else 'tesseract'
前面的准备工作完成后,让我们来运行程序吧!
# -*- coding: utf-8 -*- py3不需要
from wand.image import Image
from PIL import Image as PI
import pyocr
import pyocr.builders
import io
import sys
def main():
tools = pyocr.get_available_tools()
if len(tools) == 0:
print("No OCR tool found")
sys.exit(1)
tool = tools[0]
print("Will use tool '%s'" % (tool.get_name()))
langs = tool.get_available_languages()
print("Available languages: %s" % ", ".join(langs))
lang = langs[0]
print("Will use lang '%s'" % (lang))
req_image = []
final_text = []
image_pdf = Image(filename="./pdf_file/stackoverflow.pdf", resolution=400)
image_jpeg = image_pdf.convert('jpeg')
image_jpeg.save(filename='./pdf2img/stackoverflow.jpeg')
for img in image_jpeg.sequence:
img_page = Image(image=img)
req_image.append(img_page.make_blob('jpeg'))
for img in req_image:
txt = tool.image_to_string(
PI.open(io.BytesIO(img)),
lang=lang,
builder=pyocr.builders.TextBuilder()
)
final_text.append(txt)
print(final_text)
# for text in final_text:
# print(text)
if __name__ == '__main__':
main()
‘D:\Program Files\ImageMagick-7.0.7-Q16\CORE_RL_wand_.dll’,
‘D:\Program Files\ImageMagick-7.0.7-Q16\libMagickWand.dll’,
…
又去官网看了一下发现这段话:
Wand yet doesn’t support ImageMagick 7 which has several incompatible APIs with previous versions. For more details, see the issue #287.
卸载了之后重新安装ImageMagick-6.9.9-38-Q8-x64-dll版本问题解决。
查了半天看到一哥们也用python3.6遇到这问题,但没有解决办法,心想切换成python2.7试试。如非需要切换python版本请看一条。
(在Anaconda3中实现多版本python Spyder共存)
全程在anaconda cmd中操作
1)先在conda中创建一个名为python2的环境,并下载对应版本python2.7
conda create –name python2 python=2.7
2)激活python2环境
activate python2
3)在python2的环境下下载spyder和Jupter notebook
conda install spyder
#conda install jupyter
’ @ error/pdf.c/ReadPDFImage/
之前是在pycharm中运行的,切换成python2.7后先在spyder试了一下,结果之前的乱码变成中文了。查了一下发现是因为没有安装ghostscript,安装之后问题解决。然后我又切换为python3.6,也不报错了。
Ghostscript :https://ghostscript.com/download/gsdnld.html
(1, ‘Error opening data file \Program Files (x86)\Tesseract-OCR\tessdata/chi_sim.traineddata’)
安装Tesseract后需设置环境变量 TESSDATA_PREFIX并修改tesseract.py文件
# CHANGE THIS IF TESSERACT IS NOT IN YOUR PATH, OR IS NAMED DIFFERENTLY
#TESSERACT_CMD = 'tesseract.exe' if os.name == 'nt' else 'tesseract'
#TESSERACT_CMD = 'D:/Program Files (x86)/Tesseract-OCR/tesseract.exe' if os.name == 'nt' else 'tesseract'
TESSERACT_CMD = os.environ["TESSDATA_PREFIX"] + '/tesseract.exe' if os.name == 'nt' else 'tesseract'
D:\Program Files (x86)\Tesseract-OCR (供参考,以实际安装路径为准)
如果不设置环境变量,在安装盘之外的路径运行tesseract时会提示:
Please make sure the TESSDATA_PREFIX environment variable is set to the parent d irectory of your “tessdata” directory
注意设置环境变量后需重启项目
详见:
http://blog.csdn.net/u011389474/article/details/60139786
https://www.cnblogs.com/jamespei/p/5339769.html
pip install pdfminer3k
from pdfminer.pdfparser import PDFParser, PDFDocument
from pdfminer.pdfinterp import PDFResourceManager, PDFPageInterpreter
from pdfminer.converter import PDFPageAggregator
from pdfminer.layout import LTTextBoxHorizontal, LAParams
from pdfminer.pdfinterp import PDFTextExtractionNotAllowed
'''
解析pdf 文本,保存到txt文件中
'''
path = r'./pdf_file/stackoverflow.pdf'
def parse():
fp = open(path, 'rb') # 以二进制读模式打开
# 用文件对象来创建一个pdf文档分析器
praser = PDFParser(fp)
# 创建一个PDF文档
doc = PDFDocument()
# 连接分析器 与文档对象
praser.set_document(doc)
doc.set_parser(praser)
# 提供初始化密码
# 如果没有密码 就创建一个空的字符串
doc.initialize()
# 检测文档是否提供txt转换,不提供就忽略
if not doc.is_extractable:
raise PDFTextExtractionNotAllowed
else:
# 创建PDf 资源管理器 来管理共享资源
rsrcmgr = PDFResourceManager()
# 创建一个PDF设备对象
laparams = LAParams()
device = PDFPageAggregator(rsrcmgr, laparams=laparams)
# 创建一个PDF解释器对象
interpreter = PDFPageInterpreter(rsrcmgr, device)
# 循环遍历列表,每次处理一个page的内容
for page in doc.get_pages(): # doc.get_pages() 获取page列表
interpreter.process_page(page)
# 接受该页面的LTPage对象
layout = device.get_result()
"""
这里layout是一个LTPage对象 里面存放着 这个page解析出的各种对象 一般包括LTTextBox, LTFigure,
LTImage, LTTextBoxHorizontal 等等 想要获取文本就获得对象的text属性
"""
for x in layout:
if isinstance(x, LTTextBoxHorizontal):
with open(r'./pdf_file/stackoverflow.txt', 'a') as f:
results = x.get_text()
print(results)
f.write(results + '\n')
if __name__ == '__main__':
parse()
参考链接:
https://www.liaoxuefeng.com/wiki/001374738125095c955c1e6d8bb493182103fac9270762a000/00140767171357714f87a053a824ffd811d98a83b58ec13000
https://www.cnblogs.com/zhiyong-ITNote/p/6852113.html
http://blog.csdn.net/HuangZhang_123/article/details/61920975
https://www.cnblogs.com/wzben/p/5930538.html
https://pythontips.com/2016/02/25/ocr-on-pdf-files-using-python
http://blog.topspeedsnail.com/archives/3571
https://www.cnblogs.com/yourstars/p/5849881.html
https://segmentfault.com/q/1010000007964197