Perl使用Tesseract-OCR实现验证码识别教程

原创 2017-01-11 13:37:15 969
摘要:这篇文章主要介绍了Perl使用Tesseract-OCR实现验证码识别教程,Tesseract-OCR是一个开源图形识别引擎,需要的朋友可以参考下一、Tesseract-OCR 是什么An OCR Engine that was developed at HP Labs between 1985 and 1995… and now at Google基于Leptonica(http://lepto

这篇文章主要介绍了Perl使用Tesseract-OCR实现验证码识别教程,Tesseract-OCR是一个开源图形识别引擎,需要的朋友可以参考下

一、Tesseract-OCR 是什么
An OCR Engine that was developed at HP Labs between 1985 and 1995… and now at Google
基于Leptonica(http://leptonica.com/)图形处理库开的开源图形识别引擎。
支持Linux、Windows、Mac平台,
支持.NET、C++、Python、Java等开发语言:https://code.google.com/p/tesseract-ocr/wiki/AddOns
项目地址:https://code.google.com/p/tesseract-ocr/

二、使用方法
下载安装:https://tesseract-ocr.googlecode.com/files/tesseract-ocr-setup-3.02.02.exe
注意安装时的Path目录、数学符号、语言选项,按需选择。
执行:”tesseract yourpic.png res”
图片 yourpic.png 里面的内容会被识别后保存在 res.txt
为了更精确的识别可以去到项目地址上下载相应的各种语言的 language tessdata
例如:
简体中文 https://tesseract-ocr.googlecode.com/files/tesseract-ocr-3.02.chi_sim.tar.gz
繁体中文
下载解压后 chi_sim.traineddata 复制到 Tesseract-OCR\tessdata 即可
执行:
“tesseract yourpic.png eng” 使用 默认eng语言包
“tesseract yourpic.png sim -l chi_sim” 使用 chi_sim语言包
“tesseract yourpic.png tra -l chi_tra” 使用 chi_sim语言包
选择最接近真实数据的,方便以后修正


三、进阶使用 Training

为数不多的training tesseract-ocr中文文档

http://yy-programer.blogspot.tw/2012/08/training-tesseract-ocr-301.html
对于高精度需求的需要研究一下了,日常民用级别,默认识别加后期修正即可。


四、应用实例之吸附代理

针对 http://www.proxyfire.net/ 几个代理列表页的代理吸附

话不多说直接上代码,
pf.bat

pf.pl http://www.proxyfire.net/index.php?pageid=eliteproxylist       elite.txt
pf.pl http://www.proxyfire.net/index.php?pageid=anonymousproxylist   anony.txt
pf.pl http://www.proxyfire.net/index.php?pageid=transparentproxylist trans.txt
pf.pl http://www.proxyfire.net/index.php?pageid=socks4proxylist      s4.txt
pf.pl http://www.proxyfire.net/index.php?pageid=socks5proxylist      s5.txt
type *.txt > all.tmp
del *.txt /s/q
ren all.tmp all.txt
@pause


pf.pl

use strict;
 our $url    = $ARGV[0];
 our $file   = $ARGV[1]; 
 my $res = undef;
 my @tmp = undef;
 my @pxy = undef;     
 `wget $url -q -O ___html`;   
  open FH, "<___html";    
  @tmp = ;    
  close FH;    
  $res = join('',@tmp);    
  undef(@tmp);    
  `del ___html /s /q`; 
  @tmp = ( $res =~ /<img alt="" src="([^" border="0">]+)><\/td>(\d+)'http://www.proxyfire.net'.$tmp[$i], 'port'=>$tmp[$i+1]};    
  $i = $i + 1;} for (my $i=0; $i < @pxy; $i++) {     
  if( length(${$pxy[$i]}{ip})>0 )    {       
   `echo off & wget ${$pxy[$i]}{ip} -q -O ___png`;       
    `tesseract ___png ___ -l chi_tra`;        
     my $txt = undef;          
      open FH,"<___.txt";      
        $txt = ;       
         close FH;    
             if ( length($txt)>11 )   
                  {        
                      $txt =~ s/\s+//g;     
                             $txt =~ s/日/8/g;    
                                     $txt =~ s/昍/88/g;          
                                       $txt =~ s/s0/60/g;            
                                       $txt =~ s/s1/61/g;            
                                       $txt =~ s/s2/62/g;            
                                       $txt =~ s/s3/69/g;            
                                       $txt =~ s/s4/64/g;            
                                       $txt =~ s/s5/65/g;            
                                       $txt =~ s/s7/67/g;            
                                       $txt =~ s/s8/68/g;            
                                       $txt =~ s/s9/69/g;            
                                       $txt =~ s/0s/06/g;            
                                       $txt =~ s/1s/16/g;            
                                       $txt =~ s/2s/26/g;            
                                       
                                       $txt =~ s/3s/96/g;            
                                       $txt =~ s/4s/46/g;            
                                       $txt =~ s/5s/56/g;            
                                       $txt =~ s/6s/66/g;            
                                       $txt =~ s/7s/76/g;            
                                       $txt =~ s/8s/86/g;            
                                       $txt =~ s/9s/96/g;            
                                       $txt =~ s/ss/66/g;            
                                       $txt =~ s/\.s/\.6/g;            
                                       ${$pxy[$i]}{ip} = $txt;             
                                       my $bak1 = $txt;            
                                       my $bak2 = $txt;            
                                       $bak1 =~ s/13/19/g;            
                                       $bak1 =~ s/\.32\./\.92\./g;            
                                       $bak1 =~ s/\.33\./\.99\./g;             
                                       $bak2 =~ s/19/13/g;            
                                       $bak2 =~ s/\.243/\.249/g;            
                                       $bak2 =~ s/203\./209\./g;             
                                       open FHX,">>$file";            
                                       print FHX ${$pxy[$i]}{ip}.":".${$pxy[$i]}{port}."\n";            
                                       print FHX $bak1.":".${$pxy[$i]}{port}."\n";            
                                       print FHX $bak2.":".${$pxy[$i]}{port}."\n";            
                                       close FHX;    
                                            }     
                                               my $txt = undef;  
                                                 }}
                                                 `del ___* /s /q`;undef($url);
                                                 undef($file);
                                                 undef($res);
                                                 undef(@tmp);
                                                 undef(@pxy);

更多关于Perl使用Tesseract-OCR实现验证码识别教程请关注PHP中文网(www.php.cn)其他文章!

发布手记

热门词条