简单验证码识别 tessnet2

今天突然间对识别验证码感兴趣，于是网上搜了一下

最简单的是引用tessnet2.dll，然后通过它来识别，代码如下

 1   private void button1_Click(object sender, EventArgs e)
 2         {
 3             string strUrl = @"http://www.gz.gov.cn/sofpro/gecs/common/image.jsp?dt=Thu%20Nov%2024%202011%2017:20:21%20GMT+0800%20(China%20Standard%20Time)";
 4
 5             Bitmap image = GetSourceCode(strUrl);//识别图像
 6             pictureBox1.Image = image;
 7             tessnet2.Tesseract ocr= new tessnet2.Tesseract();//声明一个OCR类
 8             ocr.SetVariable("tessedit_char_whitelist", "0123456789"); //设置识别变量，当前只能识别数字。
 9             ocr.Init(Application.StartupPath+@"\tmpe", "eng", false); //应用当前语言包。注，Tessnet2是支持多国语的。语言包下载链接：http://code.google.com/p/tesseract-ocr/downloads/list
10             List<tessnet2.Word> result = ocr.DoOCR(image, Rectangle.Empty);//执行识别操作
11             string str="";
12             foreach (tessnet2.Word word in result) //遍历识别结果。
13                 str += word.Confidence + ":" + word.Text + Environment.NewLine;
14             MessageBox.Show(str);
15         }
16
17         private Bitmap GetSourceCode(string url)
18         {
19             WebRequest request = WebRequest.Create(url);
20             WebResponse response = request.GetResponse();
21             Stream st = response.GetResponseStream();
22             Bitmap bitmap = (Bitmap)Bitmap.FromStream(st);
23             //bitmap.Save(System.Windows.Forms.Application.StartupPath + @"\tmp.bmp", System.Drawing.Imaging.ImageFormat.Bmp);
24             //bitmap = (Bitmap)Bitmap.FromFile(System.Windows.Forms.Application.StartupPath + @"\tmp.bmp");
25             return bitmap;
26         }

由于tessnet2是在netframework2.0下生成的，所以需要配置一下config文件

<configuration>
  <startup useLegacyV2RuntimeActivationPolicy="true">
    <supportedRuntime version="v4.0" sku=".NETFramework,Version=v4.0,Profile=Client"/>
  </startup>
</configuration>

于是就可以用了

注意：http://www.gz.gov.cn/sofpro/gecs/common/image.jsp?dt=Thu%20Nov%2024%202011%2017:20:21%20GMT+0800%20(China%20Standard%20Time)是官方的一个专门是数字的一个验证码图片的网址