c#OpenCVSharp+Zxing识别条形码
参考博客：https://www.cnblogs.com/dengxiaojun/p/5278679.html，但是他的demo下载太贵了
可以下载这个https://download.csdn.net/download/dsq235612/10830805?utm_source=bbsseo，其实代码都差不多，目前只能识别简单的结构的图片
先添加引用，在nuget中添加OpenCVSharp类库和识别条码类库zxing
封装OpenCVSharp的调用代码：
public class OpencvHelper { /// <summary> /// 灰度图 /// </summary> /// <param name="srcImage">未处理的mat容器</param> /// <param name="grayImage">灰度图mat容器</param> public static void CvGrayImage(Mat srcImage, Mat grayImage) { if (srcImage.Channels() == 3) { Cv2.CvtColor(srcImage, grayImage, ColorConversionCodes.BGR2GRAY); } else { grayImage = srcImage.Clone(); } //Imshow("灰度图", grayImage); } /// <summary> /// 图像的梯度幅值 /// </summary> /// <param name="grayImage"></param> public static void CvConvertScaleAbs(Mat grayImage, Mat gradientImage) { //建立图像的梯度幅值 Mat gradientXImage = new Mat(); Mat gradientYImage = new Mat(); Cv2.Sobel(grayImage, gradientXImage, MatType.CV_32F, xorder: 1, yorder: 0, ksize: -1); Cv2.Sobel(grayImage, gradientYImage, MatType.CV_32F, xorder: 0, yorder: 1, ksize: -1); //Cv2.Scharr(grayImage, gradientXImage, MatType.CV_32F, 1, 0);//CV_16S CV_32F //Cv2.Scharr(grayImage, gradientYImage, MatType.CV_32F, 0, 1); //因为我们需要的条形码在需要X方向水平,所以更多的关注X方向的梯度幅值,而省略掉Y方向的梯度幅值 Cv2.Subtract(gradientXImage, gradientYImage, gradientImage); //归一化为八位图像 Cv2.ConvertScaleAbs(gradientImage, gradientImage); //看看得到的梯度图像是什么样子 //Imshow("图像的梯度幅值", gradientImage); } /// <summary> /// 二值化图像 /// </summary> public static void BlurImage(Mat gradientImage, Mat blurImage, Mat thresholdImage) { //对图片进行相应的模糊化,使一些噪点消除 //new OpenCvSharp.Size(12, 12); (9,9) Cv2.Blur(gradientImage, blurImage, new OpenCvSharp.Size(6, 6)); //Cv2.GaussianBlur(gradientImage, blurImage, new OpenCvSharp.Size(7, 7), 0);//Size必须是奇数 //模糊化以后进行阈值化,得到到对应的黑白二值化图像,二值化的阈值可以根据实际情况调整 Cv2.Threshold(blurImage, thresholdImage, 210, 255, ThresholdTypes.Binary); //看看二值化图像 //Imshow("二值化图像", thresholdImage); } /// <summary> /// 闭运算 /// </summary> public static void MorphImage(Mat thresholdImage, Mat morphImage) { //二值化以后的图像,条形码之间的黑白没有连接起来,就要进行形态学运算,消除缝隙,相当于小型的黑洞,选择闭运算 //因为是长条之间的缝隙,所以需要选择宽度大于长度 Mat kernel = Cv2.GetStructuringElement(MorphShapes.Rect, new OpenCvSharp.Size(21, 7)); Cv2.MorphologyEx(thresholdImage, morphImage, MorphTypes.Close, kernel); //看看形态学操作以后的图像 //Imshow("闭运算", morphImage); } /// <summary> /// 膨胀腐蚀 /// </summary> public static void DilationErosionImage(Mat morphImage) { //现在要让条形码区域连接在一起,所以选择膨胀腐蚀,而且为了保持图形大小基本不变,应该使用相同次数的膨胀腐蚀 //先腐蚀,让其他区域的亮的地方变少最好是消除,然后膨胀回来,消除干扰,迭代次数根据实际情况选择 OpenCvSharp.Size size = new OpenCvSharp.Size(3, 3); OpenCvSharp.Point point = new OpenCvSharp.Point(-1, -1); Cv2.Erode(morphImage, morphImage, Cv2.GetStructuringElement(MorphShapes.Rect, size), point, 4); Cv2.Dilate(morphImage, morphImage, Cv2.GetStructuringElement(MorphShapes.Rect, size), point, 4); //看看形态学操作以后的图像 //Imshow("膨胀腐蚀", morphImage); } /// <summary> /// 显示处理后的图片 /// </summary> /// <param name="name">处理过程名称</param> /// <param name="srcImage">图片盒子</param> public static void Imshow(string name, Mat srcImage) { using (var window = new Window(name, image: srcImage, flags: WindowMode.AutoSize)) { Cv2.WaitKey(0); } //Cv2.ImShow(name, srcImage); //Cv2.WaitKey(0); } /// <summary> /// 旋转图片 /// </summary> public static void RotateImage(Mat src, Mat dst, double angle, double scale) { var imageCenter = new Point2f(src.Cols / 2f, src.Rows / 2f); var rotationMat = Cv2.GetRotationMatrix2D(imageCenter, angle, scale); Cv2.WarpAffine(src, dst, rotationMat, src.Size()); } }
调用封装的OpenCVSharp类的方法
/// <summary> /// 读取图片 /// </summary> private void DiscernImage() { string filename = FileHelper.OpenImageFile(); if (string.IsNullOrEmpty(filename)) return; Image image = Image.FromFile(filename); picImage.Image = image; _imageFilePath = filename; } private void OpenCV() { if (string.IsNullOrEmpty(_imageFilePath)) return; Mat srcImage = new Mat(_imageFilePath, ImreadModes.Color); if (srcImage.Empty()) { return; } //图像转换为灰度图像 Mat grayImage = new Mat(); OpencvHelper.CvGrayImage(srcImage, grayImage); ShowImage("灰度图像", grayImage); //OpencvHelper.RotateImage(grayImage, grayImage, 50, 1); //OpencvHelper.Imshow("旋转", grayImage); //建立图像的梯度幅值 Mat gradientImage = new Mat(); OpencvHelper.CvConvertScaleAbs(grayImage, gradientImage); ShowImage("梯度幅值", gradientImage); //对图片进行相应的模糊化,使一些噪点消除 Mat blurImage = new Mat(); Mat thresholdImage = new Mat(); OpencvHelper.BlurImage(gradientImage, blurImage, thresholdImage); ShowImage("二值化", blurImage); //二值化以后的图像,条形码之间的黑白没有连接起来,就要进行形态学运算,消除缝隙,相当于小型的黑洞,选择闭运算 //因为是长条之间的缝隙,所以需要选择宽度大于长度 Mat morphImage = new Mat(); OpencvHelper.MorphImage(thresholdImage, morphImage); ShowImage("闭运算", morphImage); //现在要让条形码区域连接在一起,所以选择膨胀腐蚀,而且为了保持图形大小基本不变,应该使用相同次数的膨胀腐蚀 //先腐蚀,让其他区域的亮的地方变少最好是消除,然后膨胀回来,消除干扰,迭代次数根据实际情况选择 OpencvHelper.DilationErosionImage(morphImage); ShowImage("膨胀腐蚀", morphImage); Mat[] contours = new Mat[10000]; List<double> OutArray = new List<double>(); //接下来对目标轮廓进行查找,目标是为了计算图像面积 Cv2.FindContours(morphImage, out contours, OutputArray.Create(OutArray), RetrievalModes.External, ContourApproximationModes.ApproxSimple); //看看轮廓图像 //Cv2.DrawContours(srcImage, contours, -1, Scalar.Yellow); //OpencvHelper.Imshow("目标轮廓", srcImage); //计算轮廓的面积并且存放 for (int i = 0; i < OutArray.Count; i++) { OutArray[i] = contours[i].ContourArea(false); } List<string> codes = new List<string>(); int num = 0; while (num < 10) //找出10个面积最大的矩形 { //找出面积最大的轮廓 double minValue, maxValue; OpenCvSharp.Point minLoc, maxLoc; Cv2.MinMaxLoc(InputArray.Create(OutArray), out minValue, out maxValue, out minLoc, out maxLoc); //计算面积最大的轮廓的最小的外包矩形 RotatedRect minRect = Cv2.MinAreaRect(contours[maxLoc.Y]); //找到了矩形的角度,但是这是一个旋转矩形,所以还要重新获得一个外包最小矩形 Rect myRect = Cv2.BoundingRect(contours[maxLoc.Y]); //将扫描的图像裁剪下来,并保存为相应的结果,保留一些X方向的边界,所以对rect进行一定的扩张 myRect.X = myRect.X - (myRect.Width / 20); myRect.Width = (int)(myRect.Width * 1.1); //TermCriteria termc = new TermCriteria(CriteriaType.MaxIter, 1, 1); //Cv2.CamShift(srcImage, myRect, termc); //一次最大面积的 var a = contours.ToList(); a.Remove(contours[maxLoc.Y]); contours = a.ToArray(); OutArray.Remove(OutArray[maxLoc.Y]); string code = DiscernBarCode(srcImage, myRect); if(!string.IsNullOrEmpty(code)) { //Cv2.Rectangle(srcImage, myRect, new Scalar(0, 255, 255), 3, LineTypes.AntiAlias); codes.Add(code); } Cv2.Rectangle(srcImage, myRect, new Scalar(0, 255, 255), 3, LineTypes.AntiAlias); num++; if (contours.Count() <= 0) break; } Image img2 = CreateImage(srcImage); picFindContours.Image = img2; txtcodess.Text = string.Join("\r\n", codes); ////找出面积最大的轮廓 //double minValue, maxValue; //OpenCvSharp.Point minLoc, maxLoc; //Cv2.MinMaxLoc(InputArray.Create(OutArray), out minValue, out maxValue, out minLoc, out maxLoc); ////计算面积最大的轮廓的最小的外包矩形 //RotatedRect minRect = Cv2.MinAreaRect(contours[maxLoc.Y]); ////为了防止找错,要检查这个矩形的偏斜角度不能超标 ////如果超标,那就是没找到 //if (minRect.Angle < 2.0) //{ // //找到了矩形的角度,但是这是一个旋转矩形,所以还要重新获得一个外包最小矩形 // Rect myRect = Cv2.BoundingRect(contours[maxLoc.Y]); // //把这个矩形在源图像中画出来 // //Cv2.Rectangle(srcImage, myRect, new Scalar(0, 255, 255), 3, LineTypes.AntiAlias); // //看看显示效果,找的对不对 // //Imshow("裁剪图片", srcImage); // //将扫描的图像裁剪下来,并保存为相应的结果,保留一些X方向的边界,所以对rect进行一定的扩张 // myRect.X = myRect.X - (myRect.Width / 20); // myRect.Width = (int)(myRect.Width * 1.1); // Mat resultImage = new Mat(srcImage, myRect); // //OpencvHelper.Imshow("结果图片", resultImage); // Image img = CreateImage(resultImage); // picCode.Image = img; // DiscernBarcode(img); // //看看轮廓图像 // Cv2.DrawContours(srcImage, contours, -1, Scalar.Red); // //把这个矩形在源图像中画出来 // Cv2.Rectangle(srcImage, myRect, new Scalar(0, 255, 255), 3, LineTypes.AntiAlias); // Image img2 = CreateImage(srcImage); // picFindContours.Image = img2; // //string path = Path.GetDirectoryName(@g_sFilePath) + "\\Ok.png"; // //if (File.Exists(@path)) File.Delete(@path);//如果文件存在 则删除 // //if (!Cv2.ImWrite(@path, resultImage)) //} srcImage.Dispose(); } private void HandelCode(Mat srcImage, Rect myRect, Mat[] contours) { Mat resultImage = new Mat(srcImage, myRect); Image img = CreateImage(resultImage); picCode.Image = img; DiscernBarcode(img); //看看轮廓图像 Cv2.DrawContours(srcImage, contours, -1, Scalar.Red); //把这个矩形在源图像中画出来 Cv2.Rectangle(srcImage, myRect, new Scalar(0, 255, 255), 3, LineTypes.AntiAlias); //Image img2 = CreateImage(srcImage); //picFindContours.Image = img2; } private Image CreateImage(Mat resultImage) { byte[] bytes = resultImage.ToBytes(); MemoryStream ms = new MemoryStream(bytes); return Bitmap.FromStream(ms, true); } private void ShowImage(string name, Mat resultImage) { //Image img = CreateImage(resultImage); //frmShowImage frm = new frmShowImage(name, img); //frm.ShowDialog(); } /// <summary> /// 解析条形码图片 /// </summary> private string DiscernBarCode(Mat srcImage, Rect myRect) { try { Mat resultImage = new Mat(srcImage, myRect); Image img = CreateImage(resultImage); Bitmap pImg = MakeGrayscale3((Bitmap)img); BarcodeReader reader = new BarcodeReader(); reader.Options.CharacterSet = "UTF-8"; Result result = reader.Decode(new Bitmap(pImg)); Console.Write(result); if (result != null) return result.ToString(); else return ""; } catch (Exception ex) { Console.Write(ex); return ""; } } /// <summary> /// 解析条形码图片 /// </summary> private void DiscernBarcode(Image primaryImage) { //Bitmap pImg = MakeGrayscale3((Bitmap)primaryImage); picHandel.Image = primaryImage; BarcodeReader reader = new BarcodeReader(); reader.Options.CharacterSet = "UTF-8"; Result result = reader.Decode(new Bitmap(primaryImage));//Image.FromFile(path) Console.Write(result); if (result != null) txtBarCode.Text = result.ToString(); else txtBarCode.Text = ""; //watch.Start(); //watch.Stop(); //TimeSpan timeSpan = watch.Elapsed; //MessageBox.Show("扫描执行时间：" + timeSpan.TotalMilliseconds.ToString()); //using (ZBar.ImageScanner scanner = new ZBar.ImageScanner()) //{ // scanner.SetConfiguration(ZBar.SymbolType.None, ZBar.Config.Enable, 0); // scanner.SetConfiguration(ZBar.SymbolType.CODE39, ZBar.Config.Enable, 1); // scanner.SetConfiguration(ZBar.SymbolType.CODE128, ZBar.Config.Enable, 1); // List<ZBar.Symbol> symbols = new List<ZBar.Symbol>(); // symbols = scanner.Scan((Image)pImg); // if (symbols != null && symbols.Count > 0) // { // //string result = string.Empty; // //symbols.ForEach(s => result += "条码内容:" + s.Data + " 条码质量:" + s.Type + Environment.NewLine); // txtBarCode.Text = symbols.FirstOrDefault().Data; // } // else // { // txtBarCode.Text = ""; // } //} }
截图出来的条形码进行灰度处理
/// <summary> /// 处理图片灰度 /// </summary> /// <param name="original"></param> /// <returns></returns> public static Bitmap MakeGrayscale3(Bitmap original) { //create a blank bitmap the same size as original Bitmap newBitmap = new Bitmap(original.Width, original.Height); //get a graphics object from the new image Graphics g = Graphics.FromImage(newBitmap); //create the grayscale ColorMatrix System.Drawing.Imaging.ColorMatrix colorMatrix = new System.Drawing.Imaging.ColorMatrix( new float[][] { new float[] {.3f, .3f, .3f, 0, 0}, new float[] {.59f, .59f, .59f, 0, 0}, new float[] {.11f, .11f, .11f, 0, 0}, new float[] {0, 0, 0, 1, 0}, new float[] {0, 0, 0, 0, 1} }); //create some image attributes ImageAttributes attributes = new ImageAttributes(); //set the color matrix attribute attributes.SetColorMatrix(colorMatrix); //draw the original image on the new image //using the grayscale color matrix g.DrawImage(original, new Rectangle(0, 0, original.Width, original.Height), 0, 0, original.Width, original.Height, GraphicsUnit.Pixel, attributes); //dispose the Graphics object g.Dispose(); return newBitmap; }
效果图：
