文章目录

总结

1、轮廓可以绘制：正外接矩形、最小外接矩形、最小外接圆、最小椭圆：其中最小外接矩形与最小椭圆在绘制时，如果轮廓点少于5个，会导致异常；！！
2、轮廓多边形逼近函数ApproxPolyDp()：输出结果为Point数组，可以使用CvInvoke.Polylines()函数进行绘制；
3、根据轮廓的面积，大小（按比例或绝对值）进行筛选，可以有效的提取出目标轮廓；
4、实验中第二个实例使用颜色识别提取轮廓更好。


epsilon : 单位为像素

Code

using System;
using System.Collections.Generic;
using System.Linq;
using System.Text;
using System.Threading.Tasks;

using Emgu.CV;
using Emgu.Util;
using Emgu.CV.Structure;
using Emgu.CV.CvEnum;
using Emgu.CV.Util;
using System.Drawing;


namespace lesson26
{
   
    class Program
    {
   
        static void Main(string[] args)
        {
   
            最小外接圆
            //Mat src = CvInvoke.Imread("10.png");
            //Mat dst = src.Clone();
            //Mat grayImg = new Mat();
            //CvInvoke.Imshow("input", src);
            //CvInvoke.CvtColor(src, grayImg, ColorConversion.Bgr2Gray);
            //CvInvoke.Threshold(grayImg, grayImg, 100, 255, ThresholdType.Binary);

            //VectorOfVectorOfPoint contours = new VectorOfVectorOfPoint();
            //VectorOfRect hierarchy = new VectorOfRect();
            //CvInvoke.FindContours(grayImg, contours, hierarchy, RetrType.External, ChainApproxMethod.ChainApproxNone);
            //for(int i = 0; i < contours.Size; i++)
            //{
   
            // CircleF circleF = CvInvoke.MinEnclosingCircle(contours[i]); //计算最小外接圆
            // RotatedRect rotatedRect = CvInvoke.FitEllipse(contours[i]); //计算外接椭圆
            // Point rcenter = new Point((int)rotatedRect.Center.X, (int)rotatedRect.Center.Y);
            // Point center = new Point((int)circleF.Center.X,(int)circleF.Center.Y);
            // CvInvoke.Circle(dst, center, (int)circleF.Radius, new MCvScalar(255, 0, 0), 2);
            // CvInvoke.Ellipse(dst, rcenter, new Size((int)(rotatedRect.Size.Width/2), (int)(rotatedRect.Size.Height/2)), rotatedRect.Angle, 0, 360,
            // new MCvScalar(0, 0, 255), 2);
            // ///求取面积、周长、多边形逼近
            // double length = CvInvoke.ArcLength(contours[i], true); //计算轮廓周长
            // double area = CvInvoke.ContourArea(contours[i], false); //计算轮廓面积
            // Console.WriteLine("contour {0} area: {1:N2}", i, length);
            // Console.WriteLine("contour {0} length: {1:N2}", i, area);

            // VectorOfPoint approxPoly = new VectorOfPoint();
            // CvInvoke.ApproxPolyDP(contours[i], approxPoly, length * 0.02 , true); //多变形轮廓拟合
            // CvInvoke.Polylines(dst, approxPoly, true, new MCvScalar(255, 255, 0), 2); //绘制拟合多边形
            //}
            //CvInvoke.Imshow("result", dst);
            //CvInvoke.WaitKey(0);

            //利用掩膜不规则轮廓提取
            Mat src = CvInvoke.Imread("220.jpg");
            Mat dst = new Mat(src.Size, DepthType.Cv8U, 3);
            dst.SetTo(new MCvScalar(0, 0, 0));
            Mat maskImg = dst.Clone();            
            CvInvoke.Imshow("input", src);

            Mat grayImg = new Mat();
            CvInvoke.CvtColor(src, grayImg, ColorConversion.Bgr2Gray);
            CvInvoke.Threshold(grayImg, grayImg, 100, 255, ThresholdType.BinaryInv);
            CvInvoke.Imshow("threshold", grayImg);

            VectorOfVectorOfPoint contours = new VectorOfVectorOfPoint();
            VectorOfRect hierarchy = new VectorOfRect();
            CvInvoke.FindContours(grayImg, contours, hierarchy, RetrType.Tree, ChainApproxMethod.ChainApproxNone);
            for(int i = 0; i < contours.Size; i++)
            {
   
                dst.SetTo(new MCvScalar(0, 0, 0));
                maskImg.SetTo(new MCvScalar(0, 0, 0));
                CvInvoke.DrawContours(maskImg, contours, i, new MCvScalar(255, 255, 255), -1);
                double area = CvInvoke.ContourArea(contours[i], false);
                if (area > 60)          //简单的轮廓面积筛选 
                {
   
                    Mat mask = new Mat();
                    CvInvoke.CvtColor(maskImg, mask, ColorConversion.Bgr2Gray);
                    src.CopyTo(dst, mask);
                    CvInvoke.Imshow("mask", mask);
                    CvInvoke.Imshow("result", dst);
                    CvInvoke.WaitKey(0);
                }
            }
            CvInvoke.WaitKey(0);
        }
    }
}

效果

1、最小外接圆，最小外接矩形：

2、轮廓提取