C#验证码识别类完整实例

 using System; using System.Collections.Generic; using System.Linq; using System.Text; using System.Drawing; using System.Drawing.Imaging; using System.Runtime.InteropServices; namespace 验证码处理 { class VerifyCode { public Bitmap bmpobj; public VerifyCode(Bitmap pic) { bmpobj = new Bitmap(pic);  //转换为Format32bppRgb } /// 
 /// 根据RGB，计算灰度值 /// 
 /// Color值 /// 灰度值，整型 private int GetGrayNumColor(System.Drawing.Color posClr) { return (posClr.R * 19595 + posClr.G * 38469 + posClr.B * 7472) >> 16; } /// 
 /// 灰度转换,逐点方式 /// 
 public void GrayByPixels() { for (int i = 0; i  /// 去图形边框 ///  ///  public void ClearPicBorder(int borderWidth) { for (int i = 0; i  bmpobj.Width - 1 - borderWidth || i > bmpobj.Height - 1 - borderWidth) bmpobj.SetPixel(j, i, Color.FromArgb(255, 255, 255)); } } } /// 
 /// 灰度转换,逐行方式 /// 
 public void GrayByLine() { Rectangle rec = new Rectangle(0, 0, bmpobj.Width, bmpobj.Height); BitmapData bmpData = bmpobj.LockBits(rec, ImageLockMode.ReadWrite, bmpobj.PixelFormat);// PixelFormat.Format32bppPArgb); //  bmpData.PixelFormat = PixelFormat.Format24bppRgb; IntPtr scan0 = bmpData.Scan0; int len = bmpobj.Width * bmpobj.Height; int[] pixels = new int[len]; Marshal.Copy(scan0, pixels, 0, len); //对图片进行处理 int GrayValue = 0; for (int i = 0; i  /// 得到有效图形并调整为可平均分割的大小 ///  /// 灰度背景分界值 /// 有效字符数 ///  public void GetPicValidByValue(int dgGrayValue, int CharsCount) { int posx1 = bmpobj.Width; int posy1 = bmpobj.Height; int posx2 = 0; int posy2 = 0; for (int i = 0; i  j) posx1 = j; if (posy1 > i) posy1 = i; if (posx2  leftSpan) posx1 = posx1 - leftSpan; if (posx2 + Span - leftSpan  /// 得到有效图形,图形为类变量 ///  /// 灰度背景分界值 /// 有效字符数 ///  public void GetPicValidByValue(int dgGrayValue) { int posx1 = bmpobj.Width; int posy1 = bmpobj.Height; int posx2 = 0; int posy2 = 0; for (int i = 0; i  j) posx1 = j; if (posy1 > i) posy1 = i; if (posx2  /// 得到有效图形,图形由外面传入 ///  /// 灰度背景分界值 /// 有效字符数 ///  public Bitmap GetPicValidByValue(Bitmap singlepic, int dgGrayValue) { int posx1 = singlepic.Width; int posy1 = singlepic.Height; int posx2 = 0; int posy2 = 0; for (int i = 0; i  j) posx1 = j; if (posy1 > i) posy1 = i; if (posx2  /// 平均分割图片 ///  /// 水平上分割数 /// 垂直上分割数 /// 分割好的图片数组 public Bitmap [] GetSplitPics(int RowNum,int ColNum) { if (RowNum == 0 || ColNum == 0) return null; int singW = bmpobj.Width / RowNum; int singH = bmpobj.Height / ColNum; Bitmap [] PicArray=new Bitmap[RowNum*ColNum]; Rectangle cloneRect; for (int i = 0; i  /// 返回灰度图片的点阵描述字串，1表示灰点，0表示背景 ///  /// 灰度图 /// 背前景灰色界限 ///  public string GetSingleBmpCode(Bitmap singlepic, int dgGrayValue) { Color piexl; string code = ""; for (int posy = 0; posy  /// 得到灰度图像前景背景的临界值 最大类间方差法 ///  /// 前景背景的临界值 public int GetDgGrayValue() { int[] pixelNum = new int[256];   //图象直方图，共256个点 int n, n1, n2; int total;        //total为总和，累计值 double m1, m2, sum, csum, fmax, sb;  //sb为类间方差，fmax存储最大方差值 int k, t, q; int threshValue = 1;      // 阈值 //生成直方图 for (int i = 0; i  255) q = 255; total = total + pixelNum[q];  //total为总和，累计值 } pixelNum[k] = (int)((float)total / 5.0 + 0.5);  //平滑化，左边2个+中间1个+右边2个灰度，共5个，所以总和除以5，后面加0.5是用修正值 } //求阈值 sum = csum = 0.0; n = 0; //计算总的图象的点数和质量矩，为后面的计算做准备 for (k = 0; k <= 255; k++) { sum += (double)k * (double)pixelNum[k];  //x*f(x)质量矩，也就是每个灰度的值乘以其点数（归一化后为概率），sum为其总和 n += pixelNum[k];      //n为图象总的点数，归一化后就是累积概率 } fmax = -1.0;       //类间方差sb不可能为负，所以fmax初始值为-1不影响计算的进行 n1 = 0; for (k = 0; k <256; k++)     //对每个灰度（从0到255）计算一次分割后的类间方差sb { n1 += pixelNum[k];     //n1为在当前阈值遍前景图象的点数 if (n1 == 0) { continue; }    //没有分出前景后景 n2 = n - n1;       //n2为背景图象的点数 if (n2 == 0) { break; }    //n2为0表示全部都是后景图象，与n1=0情况类似，之后的遍历不可能使前景点数增加，所以此时可以退出循环 csum += (double)k * pixelNum[k];  //前景的“灰度的值*其点数”的总和 m1 = csum / n1;      //m1为前景的平均灰度 m2 = (sum - csum) / n2;    //m2为背景的平均灰度 sb = (double)n1 * (double)n2 * (m1 - m2) * (m1 - m2); //sb为类间方差 if (sb > fmax)     //如果算出的类间方差大于前一次算出的类间方差 { fmax = sb;      //fmax始终为最大类间方差（otsu） threshValue = k;    //取最大类间方差时对应的灰度的k就是最佳阈值 } } return threshValue; } /// 
 /// 去掉杂点（适合杂点/杂线粗为1） /// 
 /// 背前景灰色界限 ///  public void ClearNoise(int dgGrayValue, int MaxNearPoints) { Color piexl; int nearDots = 0; //逐点判断 for (int i = 0; i  /// 3×3中值滤波除杂 ///  ///  public void ClearNoise(int dgGrayValue) { int x, y; byte[] p = new byte[9]; //最小处理窗口3*3 byte s; //byte[] lpTemp=new BYTE[nByteWidth*nHeight]; int i, j; //--!!!!!!!!!!!!!!下面开始窗口为3×3中值滤波!!!!!!!!!!!!!!!! for (y = 1; y  p[i]) { s = p[j]; p[j] = p[i]; p[i] = s; } } } //  if (bmpobj.GetPixel(x, y).R  /// 该函数用于对图像进行腐蚀运算。结构元素为水平方向或垂直方向的三个点， /// 中间点位于原点；或者由用户自己定义3×3的结构元素。 ///  /// 前后景临界值 /// 腐蚀方式：0表示水平方向，1垂直方向，2自定义结构元素。 ///  自定义的3×3结构元素 public void ErosionPic(int dgGrayValue, int nMode, bool[,] structure) { int lWidth = bmpobj.Width; int lHeight = bmpobj.Height; Bitmap newBmp = new Bitmap(lWidth, lHeight); int i, j, n, m;    //循环变量 if (nMode == 0) { //使用水平方向的结构元素进行腐蚀 // 由于使用1×3的结构元素，为防止越界，所以不处理最左边和最右边 // 的两列像素 for (j = 0; j  dgGrayValue || bmpobj.GetPixel(i, j).R > dgGrayValue || bmpobj.GetPixel(i + 1, j).R > dgGrayValue) newBmp.SetPixel(i, j, Color.White); } } } else if (nMode == 1) { //使用垂真方向的结构元素进行腐蚀 // 由于使用3×1的结构元素，为防止越界，所以不处理最上边和最下边 // 的两行像素 for (j = 1; j  dgGrayValue || bmpobj.GetPixel(i, j).R > dgGrayValue || bmpobj.GetPixel(i, j + 1).R > dgGrayValue) newBmp.SetPixel(i, j, Color.White); } } } else { if (structure.Length != 9) //检查自定义结构 return; //使用自定义的结构元素进行腐蚀 // 由于使用3×3的结构元素，为防止越界，所以不处理最左边和最右边 // 的两列像素和最上边和最下边的两列像素 for (j = 1; j  dgGrayValue) { newBmp.SetPixel(i, j, Color.White); break; } } } } } } bmpobj = newBmp; } /// 
 /// 该函数用于对图像进行细化运算。要求目标图像为灰度图像 /// 
 ///  public void ThiningPic(int dgGrayValue) { int lWidth = bmpobj.Width; int lHeight = bmpobj.Height; // Bitmap newBmp = new Bitmap(lWidth, lHeight); bool bModified;    //脏标记 int i, j, n, m;    //循环变量 //四个条件 bool bCondition1; bool bCondition2; bool bCondition3; bool bCondition4; int nCount;  //计数器 int[,] neighbour = new int[5, 5];  //5×5相邻区域像素值 bModified = true; while (bModified) { bModified = false; //由于使用5×5的结构元素，为防止越界，所以不处理外围的几行和几列像素 for (j = 2; j  dgGrayValue) { if (bmpobj.GetPixel(i, j).R <255) bmpobj.SetPixel(i, j, Color.White); continue; } //获得当前点相邻的5×5区域内像素值，白色用0代表，黑色用1代表 for (m = 0; m <5; m++) { for (n = 0; n <5; n++) { neighbour[m, n] = bmpobj.GetPixel(i + m - 2, j + n - 2).R = 2 && nCount <= 6) { bCondition1 = true; } //判断Z0(P1)=1 nCount = 0; if (neighbour[1, 2] == 0 && neighbour[1, 1] == 1) nCount++; if (neighbour[1, 1] == 0 && neighbour[2, 1] == 1) nCount++; if (neighbour[2, 1] == 0 && neighbour[3, 1] == 1) nCount++; if (neighbour[3, 1] == 0 && neighbour[3, 2] == 1) nCount++; if (neighbour[3, 2] == 0 && neighbour[3, 3] == 1) nCount++; if (neighbour[3, 3] == 0 && neighbour[2, 3] == 1) nCount++; if (neighbour[2, 3] == 0 && neighbour[1, 3] == 1) nCount++; if (neighbour[1, 3] == 0 && neighbour[1, 2] == 1) nCount++; if (nCount == 1) bCondition2 = true; //判断P2*P4*P8=0 or Z0(p2)!=1 if (neighbour[1, 2] * neighbour[2, 1] * neighbour[2, 3] == 0) { bCondition3 = true; } else { nCount = 0; if (neighbour[0, 2] == 0 && neighbour[0, 1] == 1) nCount++; if (neighbour[0, 1] == 0 && neighbour[1, 1] == 1) nCount++; if (neighbour[1, 1] == 0 && neighbour[2, 1] == 1) nCount++; if (neighbour[2, 1] == 0 && neighbour[2, 2] == 1) nCount++; if (neighbour[2, 2] == 0 && neighbour[2, 3] == 1) nCount++; if (neighbour[2, 3] == 0 && neighbour[1, 3] == 1) nCount++; if (neighbour[1, 3] == 0 && neighbour[0, 3] == 1) nCount++; if (neighbour[0, 3] == 0 && neighbour[0, 2] == 1) nCount++; if (nCount != 1) bCondition3 = true; } //判断P2*P4*P6=0 or Z0(p4)!=1 if (neighbour[1, 2] * neighbour[2, 1] * neighbour[3, 2] == 0) { bCondition4 = true; } else { nCount = 0; if (neighbour[1, 1] == 0 && neighbour[1, 0] == 1) nCount++; if (neighbour[1, 0] == 0 && neighbour[2, 0] == 1) nCount++; if (neighbour[2, 0] == 0 && neighbour[3, 0] == 1) nCount++; if (neighbour[3, 0] == 0 && neighbour[3, 1] == 1) nCount++; if (neighbour[3, 1] == 0 && neighbour[3, 2] == 1) nCount++; if (neighbour[3, 2] == 0 && neighbour[2, 2] == 1) nCount++; if (neighbour[2, 2] == 0 && neighbour[1, 2] == 1) nCount++; if (neighbour[1, 2] == 0 && neighbour[1, 1] == 1) nCount++; if (nCount != 1) bCondition4 = true; } if (bCondition1 && bCondition2 && bCondition3 && bCondition4) { bmpobj.SetPixel(i, j, Color.White); bModified = true; } else { bmpobj.SetPixel(i, j, Color.Black); } } } } // 复制细化后的图像 //  bmpobj = newBmp; } /// 
 /// 锐化要启用不安全代码编译 /// 
 /// 锐化程度。取值[0,1]。值越大锐化程度越高 /// 锐化后的图像 public void Sharpen(float val) { int w = bmpobj.Width; int h = bmpobj.Height; Bitmap bmpRtn = new Bitmap(w, h, PixelFormat.Format24bppRgb); BitmapData srcData = bmpobj.LockBits(new Rectangle(0, 0, w, h), ImageLockMode.ReadOnly, PixelFormat.Format24bppRgb); BitmapData dstData = bmpRtn.LockBits(new Rectangle(0, 0, w, h), ImageLockMode.WriteOnly, PixelFormat.Format24bppRgb); unsafe { byte* pIn = (byte*)srcData.Scan0.ToPointer(); byte* pOut = (byte*)dstData.Scan0.ToPointer(); int stride = srcData.Stride; byte* p; for (int y = 0; y  0) { vR = Math.Min(255, vR); } else { vR = Math.Max(0, vR); } if (vG > 0) { vG = Math.Min(255, vG); } else { vG = Math.Max(0, vG); } if (vB > 0) { vB = Math.Min(255, vB); } else { vB = Math.Max(0, vB); } pOut[0] = (byte)vB; pOut[1] = (byte)vG; pOut[2] = (byte)vR; } pIn += 3; pOut += 3; }// end of x pIn += srcData.Stride - w * 3; pOut += srcData.Stride - w * 3; } // end of y } bmpobj.UnlockBits(srcData); bmpRtn.UnlockBits(dstData); bmpobj = bmpRtn; } /// 
 /// 图片二值化 /// 
 ///  public void BitmapTo1Bpp(Double hsb) { int w = bmpobj.Width; int h = bmpobj.Height; Bitmap bmp = new Bitmap(w, h, PixelFormat.Format1bppIndexed); BitmapData data = bmp.LockBits(new Rectangle(0, 0, w, h), ImageLockMode.ReadWrite, PixelFormat.Format1bppIndexed); for (int y = 0; y = hsb) scan[x / 8] |= (byte)(0x80 >> (x % 8)); } Marshal.Copy(scan, 0, (IntPtr)((int)data.Scan0 + data.Stride * y), scan.Length); } bmp.UnlockBits(data); bmpobj = bmp; } } }