请求大侠帮忙把这段JS改写成ASP.NET版本~

Question

var image = document.querySelector(".greyfont2 img");       //如果要用在greasemonkey脚本里,可以把下面的代码放在image的onload事件里          
var canvas = document.createElement('canvas');                 
var ctx = canvas.getContext("2d");                 
var numbers = [                           //模板,依次是0-9十个数字对应的明暗值字符串
"111000111100000001100111001001111100001111100001111100001111100001111100001111100001111100100111001100000001111000111111111111111111111111111111",
"111000111100000111100000111111100111111100111111100111111100111111100111111100111111100111111100111100000000100000000111111111111111111111111111",
"100000111000000011011111001111111001111111001111110011111100111111001111110011111100111111001111111000000001000000001111111111111111111111111111",
"100000111000000001011111001111111001111110011100000111100000011111110001111111001111111001011110001000000011100000111111111111111111111111111111",
"111110011111100011111100011111000011110010011110010011100110011100110011000000000000000000111110011111110011111110011111111111111111111111111111",
"000000001000000001001111111001111111001111111000001111000000011111110001111111001111111001011110001000000011100000111111111111111111111111111111",
"111000011110000001100111101100111111001111111001000011000000001000111000001111100001111100100111000100000001111000011111111111111111111111111111",
"100000000100000000111111100111111101111111001111110011111110111111100111111101111111001111111001111110011111110011111111111111111111111111111111",
"110000011100000001100111001100111001100011011110000011110000011100110001001111100001111100000111000100000001110000011111111111111111111111111111",
"110000111100000001000111001001111100001111100000111000100000000110000100111111100111111001101111001100000011110000111111111111111111111111111111"];
var captcha = "";                         //存放识别后的验证码
canvas.width = image.width;
canvas.height = image.height;
document.body.appendChild(canvas);
ctx.drawImage(image, 0, 0);
for (var i = 0; i < 4; i++) {
    var pixels = ctx.getImageData(13 * i + 7, 3, 9, 16).data;
    var ldString = "";
    for (var j = 0,length = pixels.length; j < length; j += 4) {
        ldString = ldString + (+(pixels[j] * 0.3 + pixels[j + 1] * 0.59 + pixels[j + 2] * 0.11 >= 140));
    }
    var comms = numbers.map(function (value) {                      //为了100%识别率,这里不能直接判断是否和模板字符串相等,因为可能有个别0被计算成1,或者相反
        return ldString.split("").filter(function (v, index) {
            return value[index] === v
        }).length
    });
    captcha += comms.indexOf(Math.max.apply(null, comms));          //添加到识别好的验证码中
}
document.querySelector("input[name=validateCode]").value = captcha; //写入目标文本框麻烦把这段代码改写成ASP.NET版本的，麻烦各位大侠了！

Answer 1

这个是识别验证码的代码， 是JS版本的，但我需要把他改成ASP.NET 做成DLL 用，

Answer 2

就是 注册  登录时 使用的验证码吗？？  我有一段生成验证码的代码
using System;
using System.Drawing;
using System.Drawing.Imaging;
using System.Web.UI;
using System.Drawing.Drawing2D;
using System.IO;
namespace Validate
{
    /// <summary>
    /// 生成验证码的类
    /// </summary>
    public class ValidateNumber
    {
        public ValidateNumber()
        {
        }
        /// <summary>
        /// 验证码的最大长度
        /// </summary>
        public int MaxLength
        {
            get { return 10; }
        }
        /// <summary>
        /// 验证码的最小长度
        /// </summary>
        public int MinLength
        {
            get { return 1; }
        }
        /// <summary>
        /// 生成验证码
        /// </summary>
        /// <param name="length">指定验证码的长度</param>
        /// <returns></returns>
        public string CreateValidateNumber(int length)
        {
            int[] randMembers = new int[length];   //随机生成的数字集合
            int[] validateNums = new int[length];  //抽取出来的随机数字集合
            string validateNumberStr = "";
            //生成起始序列值
            int seekSeek = unchecked((int)DateTime.Now.Ticks);
            Random seekRand = new Random(seekSeek);
            int beginSeek = (int)seekRand.Next(0, Int32.MaxValue - length * 10000);
            int[] seeks = new int[length];
            for (int i = 0; i < length; i++)
            {
                beginSeek += 10000;
                seeks[i] = beginSeek;
            }
            //生成随机数字
            for (int i = 0; i < length; i++)
            {
                Random rand = new Random(seeks[i]);
                int pownum = 1 * (int)Math.Pow(10, length);
                randMembers[i] = rand.Next(pownum, Int32.MaxValue);
            }
            //抽取随机数字
            for (int i = 0; i < length; i++)
            {
                string numStr = randMembers[i].ToString();
                int numLength = numStr.Length;
                Random rand = new Random();
                int numPosition = rand.Next(0, numLength - 1);
                validateNums[i] = Int32.Parse(numStr.Substring(numPosition, 1));
            }
            //生成验证码
            for (int i = 0; i < length; i++)
            {
                validateNumberStr += validateNums[i].ToString();
            }
            return validateNumberStr;
        }
        /// <summary>
        /// 创建验证码的图片
        /// </summary>
        /// <param name="containsPage">要输出到的page对象</param>
        /// <param name="validateNum">验证码</param>
        public void CreateValidateGraphic(Page containsPage, string validateNum)
        {
            Bitmap image = new Bitmap((int)Math.Ceiling(validateNum.Length * 12.5), 22);
            Graphics g = Graphics.FromImage(image);
            try
            {
                //生成随机生成器
                Random random = new Random();
                //清空图片背景色
                g.Clear(Color.White);
                //画图片的干扰线
                for (int i = 0; i < 25; i++)
                {
                    int x1 = random.Next(image.Width);
                    int x2 = random.Next(image.Width);
                    int y1 = random.Next(image.Height);
                    int y2 = random.Next(image.Height);
                    g.DrawLine(new Pen(Color.Silver), x1, y1, x2, y2);
                }
                Font font = new Font("Arial", 12, (FontStyle.Bold | FontStyle.Italic));
                LinearGradientBrush brush = new LinearGradientBrush(new Rectangle(0, 0, image.Width, image.Height),
                 Color.Blue, Color.DarkRed, 1.2f, true);
                g.DrawString(validateNum, font, brush, 3, 2);
                //画图片的前景干扰点
                for (int i = 0; i < 100; i++)
                {
                    int x = random.Next(image.Width);
                    int y = random.Next(image.Height);
                    image.SetPixel(x, y, Color.FromArgb(random.Next()));
                }
                //画图片的边框线
                g.DrawRectangle(new Pen(Color.Silver), 0, 0, image.Width - 1, image.Height - 1);
                //保存图片数据
                MemoryStream stream = new MemoryStream();
                image.Save(stream, ImageFormat.Jpeg);
                //输出图片
                containsPage.Response.Clear();
                containsPage.Response.ContentType = "image/jpeg";
                containsPage.Response.BinaryWrite(stream.ToArray());
            }
            finally
            {
                g.Dispose();
                image.Dispose();
            }
        }
        /// <summary>
        /// 得到验证码图片的长度
        /// </summary>
        /// <param name="validateNumLength">验证码的长度</param>
        /// <returns></returns>
        public static int GetImageWidth(int validateNumLength)
        {
            return (int)(validateNumLength * 12.5);
        }
        /// <summary>
        /// 得到验证码的高度
        /// </summary>
        /// <returns></returns>
        public static double GetImageHeight()
        {
            return 22.5;
        }
    }
}

Answer 3

这不好改吧
既然知道你的需求了，识别验证码，那么去百度一个c#版本的不就行了么？
纯用C#写了CSDN验证码识别程序！

Answer 4

以前研究了一下  你那个也就识别最简单的验证码（不带杂点、噪音线、没做过扭曲）  仅仅做了个灰度转换和分割  这边有个识别验证码的类
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
        }        /// <summary>
        /// 根据RGB，计算灰度值
        /// </summary>
        /// <param name="posClr">Color值</param>
        /// <returns>灰度值，整型</returns>
        private int GetGrayNumColor(System.Drawing.Color posClr)
        {
            return (posClr.R * 19595 + posClr.G * 38469 + posClr.B * 7472) >> 16;
        }        /// <summary>
        /// 灰度转换,逐点方式
        /// </summary>
        public void GrayByPixels()
        {
            for (int i = 0; i < bmpobj.Height; i++)
            {
                for (int j = 0; j < bmpobj.Width; j++)
                {
                    int tmpValue = GetGrayNumColor(bmpobj.GetPixel(j, i));
                    bmpobj.SetPixel(j, i, Color.FromArgb(tmpValue, tmpValue, tmpValue));
                }
            }
        }        /// <summary>
        /// 去图形边框
        /// </summary>
        /// <param name="borderWidth"></param>
        public void ClearPicBorder(int borderWidth)
        {
            for (int i = 0; i < bmpobj.Height; i++)
            {
                for (int j = 0; j < bmpobj.Width; j++)
                {
                    if (i < borderWidth || j < borderWidth || j > bmpobj.Width - 1 - borderWidth || i > bmpobj.Height - 1 - borderWidth)
                        bmpobj.SetPixel(j, i, Color.FromArgb(255, 255, 255));
                }
            }
        }        /// <summary>
        /// 灰度转换,逐行方式
        /// </summary>
        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 < len; i++)
            {
                GrayValue = GetGrayNumColor(Color.FromArgb(pixels[i]));
                pixels[i] = (byte)(Color.FromArgb(GrayValue, GrayValue, GrayValue)).ToArgb();      //Color转byte
            }            bmpobj.UnlockBits(bmpData);            ////输出
            //GCHandle gch = GCHandle.Alloc(pixels, GCHandleType.Pinned);
            //bmpOutput = new Bitmap(bmpobj.Width, bmpobj.Height, bmpData.Stride, bmpData.PixelFormat, gch.AddrOfPinnedObject());
            //gch.Free();
        }        /// <summary>
        /// 得到有效图形并调整为可平均分割的大小
        /// </summary>
        /// <param name="dgGrayValue">灰度背景分界值</param>
        /// <param name="CharsCount">有效字符数</param>
        /// <returns></returns>
        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 < bmpobj.Height; i++)      //找有效区
            {
                for (int j = 0; j < bmpobj.Width; j++)
                {
                    int pixelValue = bmpobj.GetPixel(j, i).R;
                    if (pixelValue < dgGrayValue)     //根据灰度值
                    {
                        if (posx1 > j) posx1 = j;
                        if (posy1 > i) posy1 = i;                        if (posx2 < j) posx2 = j;
                        if (posy2 < i) posy2 = i;
                    };
                };
            };
            // 确保能整除
            int Span = CharsCount - (posx2 - posx1 + 1) % CharsCount;   //可整除的差额数
            if (Span < CharsCount)
            {
                int leftSpan = Span / 2;    //分配到左边的空列 ，如span为单数,则右边比左边大1
                if (posx1 > leftSpan)
                    posx1 = posx1 - leftSpan;
                if (posx2 + Span - leftSpan < bmpobj.Width)
                    posx2 = posx2 + Span - leftSpan;
            }
            //复制新图
            Rectangle cloneRect = new Rectangle(posx1, posy1, posx2 - posx1 + 1, posy2 - posy1 + 1);
            bmpobj = bmpobj.Clone(cloneRect, bmpobj.PixelFormat);
        }        /// <summary>
        /// 得到有效图形,图形为类变量
        /// </summary>
        /// <param name="dgGrayValue">灰度背景分界值</param>
        /// <param name="CharsCount">有效字符数</param>
        /// <returns></returns>
        public void GetPicValidByValue(int dgGrayValue)
        {
            int posx1 = bmpobj.Width; int posy1 = bmpobj.Height;
            int posx2 = 0; int posy2 = 0;
            for (int i = 0; i < bmpobj.Height; i++)      //找有效区
            {
                for (int j = 0; j < bmpobj.Width; j++)
                {
                    int pixelValue = bmpobj.GetPixel(j, i).R;
                    if (pixelValue < dgGrayValue)     //根据灰度值
                    {
                        if (posx1 > j) posx1 = j;
                        if (posy1 > i) posy1 = i;                        if (posx2 < j) posx2 = j;
                        if (posy2 < i) posy2 = i;
                    };
                };
            };
            //复制新图
            Rectangle cloneRect = new Rectangle(posx1, posy1, posx2 - posx1 + 1, posy2 - posy1 + 1);
            bmpobj = bmpobj.Clone(cloneRect, bmpobj.PixelFormat);
        }        /// <summary>
        /// 得到有效图形,图形由外面传入
        /// </summary>
        /// <param name="dgGrayValue">灰度背景分界值</param>
        /// <param name="CharsCount">有效字符数</param>
        /// <returns></returns>
        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 < singlepic.Height; i++)      //找有效区
            {
                for (int j = 0; j < singlepic.Width; j++)
                {
                    int pixelValue = singlepic.GetPixel(j, i).R;
                    if (pixelValue < dgGrayValue)     //根据灰度值
                    {
                        if (posx1 > j) posx1 = j;
                        if (posy1 > i) posy1 = i;                        if (posx2 < j) posx2 = j;
                        if (posy2 < i) posy2 = i;
                    };
                };
            };
            //复制新图
            Rectangle cloneRect = new Rectangle(posx1, posy1, posx2 - posx1 + 1, posy2 - posy1 + 1);
            return singlepic.Clone(cloneRect, singlepic.PixelFormat);
        }        /// <summary>
        /// 平均分割图片
        /// </summary>
        /// <param name="RowNum">水平上分割数</param>
        /// <param name="ColNum">垂直上分割数</param>
        /// <returns>分割好的图片数组</returns>
        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 < ColNum; i++)      //找有效区
            {
                for (int j = 0; j < RowNum; j++)
                {
                    cloneRect = new Rectangle(j * singW, i * singH, singW, singH);
                    PicArray[i * RowNum + j] = bmpobj.Clone(cloneRect, bmpobj.PixelFormat);//复制小块图
                }
            }
            return PicArray;
        }

Answer 5

 
  /// <summary>
        /// 返回灰度图片的点阵描述字串，1表示灰点，0表示背景
        /// </summary>
        /// <param name="singlepic">灰度图</param>
        /// <param name="dgGrayValue">背前景灰色界限</param>
        /// <returns></returns>
        public string GetSingleBmpCode(Bitmap singlepic, int dgGrayValue)
        {
            Color piexl;
            string code = "";
            for (int posy = 0; posy < singlepic.Height; posy++)
                for (int posx = 0; posx < singlepic.Width; posx++)
                {
                    piexl = singlepic.GetPixel(posx, posy);
                    if (piexl.R < dgGrayValue)    // Color.Black )
                        code = code + "1";
                    else
                        code = code + "0";
                }
            return code;
        }        /// <summary>
        /// 得到灰度图像前景背景的临界值 最大类间方差法
        /// </summary>
        /// <returns>前景背景的临界值</returns>
        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 < bmpobj.Width; i++)
            {
                for (int j = 0; j < bmpobj.Height; j++)
                {
                    //返回各个点的颜色，以RGB表示
                    pixelNum[bmpobj.GetPixel(i, j).R]++;            //相应的直方图加1
                }
            }
            //直方图平滑化
            for (k = 0; k <= 255; k++)
            {
                total = 0;
                for (t = -2; t <= 2; t++)              //与附近2个灰度做平滑化，t值应取较小的值
                {
                    q = k + t;
                    if (q < 0)                     //越界处理
                        q = 0;
                    if (q > 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;
        }
        /// <summary>
        ///  去掉杂点（适合杂点/杂线粗为1）
        /// </summary>
        /// <param name="dgGrayValue">背前景灰色界限</param>
        /// <returns></returns>
        public void ClearNoise(int dgGrayValue, int MaxNearPoints)
        {
            Color piexl;
            int nearDots = 0;
            //逐点判断
            for (int i = 0; i < bmpobj.Width; i++)
                for (int j = 0; j < bmpobj.Height; j++)
                {
                    piexl = bmpobj.GetPixel(i, j);
                    if (piexl.R < dgGrayValue)
                    {
                        nearDots = 0;
                        //判断周围8个点是否全为空
                        if (i == 0 || i == bmpobj.Width - 1 || j == 0 || j == bmpobj.Height - 1)  //边框全去掉
                        {
                            bmpobj.SetPixel(i, j, Color.FromArgb(255, 255, 255));
                        }
                        else
                        {
                            if (bmpobj.GetPixel(i - 1, j - 1).R < dgGrayValue) nearDots++;
                            if (bmpobj.GetPixel(i, j - 1).R < dgGrayValue) nearDots++;
                            if (bmpobj.GetPixel(i + 1, j - 1).R < dgGrayValue) nearDots++;
                            if (bmpobj.GetPixel(i - 1, j).R < dgGrayValue) nearDots++;
                            if (bmpobj.GetPixel(i + 1, j).R < dgGrayValue) nearDots++;
                            if (bmpobj.GetPixel(i - 1, j + 1).R < dgGrayValue) nearDots++;
                            if (bmpobj.GetPixel(i, j + 1).R < dgGrayValue) nearDots++;
                            if (bmpobj.GetPixel(i + 1, j + 1).R < dgGrayValue) nearDots++;
                        }                        if (nearDots < MaxNearPoints)
                            bmpobj.SetPixel(i, j, Color.FromArgb(255, 255, 255));   //去掉单点 && 粗细小3邻边点
                    }
                    else  //背景
                        bmpobj.SetPixel(i, j, Color.FromArgb(255, 255, 255));
                }
        }        /// <summary>
        /// 3×3中值滤波除杂
        /// </summary>
        /// <param name="dgGrayValue"></param>
        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 < bmpobj.Height - 1; y++) //--第一行和最后一行无法取窗口
            {
                for (x = 1; x < bmpobj.Width - 1; x++)
                {
                    //取9个点的值
                    p[0] = bmpobj.GetPixel(x - 1, y - 1).R;
                    p[1] = bmpobj.GetPixel(x, y - 1).R;
                    p[2] = bmpobj.GetPixel(x + 1, y - 1).R;
                    p[3] = bmpobj.GetPixel(x - 1, y).R;
                    p[4] = bmpobj.GetPixel(x, y).R;
                    p[5] = bmpobj.GetPixel(x + 1, y).R;
                    p[6] = bmpobj.GetPixel(x - 1, y + 1).R;
                    p[7] = bmpobj.GetPixel(x, y + 1).R;
                    p[8] = bmpobj.GetPixel(x + 1, y + 1).R;
                    //计算中值
                    for (j = 0; j < 5; j++)
                    {
                        for (i = j + 1; i < 9; i++)
                        {
                            if (p[j] > p[i])
                            {
                                s = p[j];
                                p[j] = p[i];
                                p[i] = s;
                            }
                        }
                    }
                    //      if (bmpobj.GetPixel(x, y).R < dgGrayValue)
                    bmpobj.SetPixel(x, y, Color.FromArgb(p[4], p[4], p[4]));    //给有效值付中值
                }
            }
        }

Answer 6

不贴了 贴不下 上几个星期不是有个精华帖吗？ 好像是人工智能应用于识别验证码的
这里还有个扭曲验证码的类
 public class Twist
    {
        /// <summary>
        /// 正弦曲线Wave扭曲图片   
        /// </summary>
        /// <param name="srcBmp">图片路径</param>
        /// <param name="bXDir">沿Y轴扭曲则选择为True</param>
        /// <param name="dMultValue">波形的幅度倍数，越大扭曲的程度越高，一般为3</param>
        /// <param name="dPhase">波形的起始相位，取值区间[0-2*PI)</param>
        /// <returns>扭曲后的位图</returns>
        public Bitmap TwistBitmap(Bitmap srcBmp, bool bXDir, double dMultValue, double dPhase)
        {
            int w = srcBmp.Width;
            int h = srcBmp.Height;
            System.Drawing.Bitmap destBmp = new Bitmap(w, h);
            using (System.Drawing.Graphics graph = System.Drawing.Graphics.FromImage(destBmp))
            {
                graph.Clear(Color.White);
            }
            double dBaseAxisLen = bXDir ? (double)h : (double)w;
            BitmapData destData = destBmp.LockBits(new Rectangle(0, 0, w, h), ImageLockMode.WriteOnly, srcBmp.PixelFormat);
            BitmapData srcData = srcBmp.LockBits(new Rectangle(0, 0, w, h), ImageLockMode.ReadOnly, srcBmp.PixelFormat);
             //调式一定注意PixelFormat是多少，别处都没有解释的。如果是24下面指针和跨度就3倍，要是32就是4倍。
            unsafe
            {
                byte* p = (byte*)srcData.Scan0;
                byte* p2 = (byte*)destData.Scan0;
                for (int j = 0; j < h; j++)
                {
                    for (int i = 0; i < w; i++)
                    {
                        double dx = 0;
                        dx = bXDir ? (Math.PI * (double)j * 2) / dBaseAxisLen : (Math.PI * (double)i * 2) / dBaseAxisLen;
                        dx += dPhase;
                        double dy = Math.Sin(dx);
                        // 取得当前点  
                        int nOldX = 0, nOldY = 0;
                        nOldX = bXDir ? i + (int)(dy * dMultValue) : i;
                        nOldY = bXDir ? j : j + (int)(dy * dMultValue);                        if ((nOldX >= 0 && nOldX < w) && (nOldY >= 0 && nOldY < h))
                        {
                            p2[(nOldY * destData.Stride) + (nOldX * 4)] = p[(j * srcData.Stride) + (i * 4)];
                            p2[(nOldY * destData.Stride) + (nOldX * 4) + 1] = p[(j * srcData.Stride) + (i * 4) + 1];
                            p2[(nOldY * destData.Stride) + (nOldX * 4) + 2] = p[(j * srcData.Stride) + (i * 4) + 2];
                        }
                    }
                }
            }
            destBmp.UnlockBits(destData);
            srcBmp.UnlockBits(srcData);
            if (srcBmp != null)
                srcBmp.Dispose();
            return destBmp;
        }
    }

Answer 7

贴完睡觉  这只是一般的验证码  对于那些复杂的验证码还是得研究别人用什么算法生成的  你才可以试着去识别          /// <summary>
        /// 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。
        /// </summary>
        /// <param name="dgGrayValue">前后景临界值</param>
        /// <param name="nMode">腐蚀方式：0表示水平方向，1垂直方向，2自定义结构元素。</param>
        /// <param name="structure"> 自定义的3×3结构元素</param>
        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;                   if (nMode == 0)
            {
                for (j = 0; j < lHeight; j++)
                {
                    for (i = 1; i < lWidth - 1; i++)
                    {
                        //目标图像中的当前点先赋成黑色
                        newBmp.SetPixel(i, j, Color.Black);                        //如果源图像中当前点自身或者左右有一个点不是黑色，
                        //则将目标图像中的当前点赋成白色
                        if (bmpobj.GetPixel(i - 1, j).R > 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 < lHeight - 1; j++)
                {
                    for (i = 0; i < lWidth; i++)
                    {
                        //目标图像中的当前点先赋成黑色
                        newBmp.SetPixel(i, j, Color.Black);                        //如果源图像中当前点自身或者左右有一个点不是黑色，
                        //则将目标图像中的当前点赋成白色
                        if (bmpobj.GetPixel(i, j - 1).R > 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;
                for (j = 1; j < lHeight - 1; j++)
                {
                    for (i = 1; i < lWidth - 1; i++)
                    {
                        //目标图像中的当前点先赋成黑色
                        newBmp.SetPixel(i, j, Color.Black);
                        //如果原图像中对应结构元素中为黑色的那些点中有一个不是黑色，
                        //则将目标图像中的当前点赋成白色
                        for (m = 0; m < 3; m++)
                        {
                            for (n = 0; n < 3; n++)
                            {
                                if (!structure[m, n])
                                    continue;
                                if (bmpobj.GetPixel(i + m - 1, j + n - 1).R > dgGrayValue)
                                {
                                    newBmp.SetPixel(i, j, Color.White);
                                    break;
                                }
                            }
                        }
                    }
                }
            }
            bmpobj = newBmp;
        }
        /// <summary>
        /// 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
        /// </summary>
        /// <param name="dgGrayValue"></param>
        public void ThiningPic(int dgGrayValue)
        {
            int lWidth = bmpobj.Width;
            int lHeight = bmpobj.Height;
            bool bModified;            
            int i, j, n, m;                       //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 < lHeight - 2; j++)
                {
                    for (i = 2; i < lWidth - 2; i++)
                    {
                        bCondition1 = false;
                        bCondition2 = false;
                        bCondition3 = false;
                        bCondition4 = false;                        if (bmpobj.GetPixel(i, j).R > 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 < dgGrayValue ? 1 : 0;
                            }
                        }                        //逐个判断条件。
                        //判断2<=NZ(P1)<=6
                        nCount = neighbour[1, 1] + neighbour[1, 2] + neighbour[1, 3]
                               + neighbour[2, 1] + neighbour[2, 3] +
                                +neighbour[3, 1] + neighbour[3, 2] + neighbour[3, 3];
                        if (nCount >= 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;
        }

Answer 8

最近觉得写代码都没有什么意义！ 人工智能挺不错的  不过最近也在搞一个MMAS算法