图像分割 Image Segmentation

               

http://www.cnblogs.com/xrwang/archive/2010/02/28/ImageSegmentation.html?login=1

作者：王先荣
前言
    图像分割指的是将数字图像细分为多个图像子区域的过程，在OpenCv中实现了三种跟图像分割相关的算法，它们分别是：分水岭分割算法、金字塔分割算法以及均值漂移分割算法。它们的使用过程都很简单，下面的文章权且用于记录，并使该系列保持完整吧。
分水岭分割算法
    分水岭分割算法需要您或者先前算法提供标记，该标记用于指定哪些大致区域是目标，哪些大致区域是背景等等；分水岭分割算法的分割效果严重依赖于提供的标记。OpenCv中的函数cvWatershed实现了该算法，函数定义如下：




void cvWatershed(const CvArr * image, CvArr * markers)





其中：image为8为三通道的彩色图像；
      markers是单通道整型图像，它用不同的正整数来标记不同的区域，下面的代码演示了如果响应鼠标事件，并生成标记图像。



图像分割 Image Segmentation生成标记图像


图像分割 Image Segmentation





            //当鼠标按下并在源图像上移动时，在源图像上绘制分割线条        private void pbSource_MouseMove(object sender, MouseEventArgs e)        {            //如果按下了左键            if (e.Button == MouseButtons.Left)            {                if (previousMouseLocation.X >= 0 && previousMouseLocation.Y >= 0)                {                    Point p1 = new Point((int)(previousMouseLocation.X * xScale), (int)(previousMouseLocation.Y * yScale));                    Point p2 = new Point((int)(e.Location.X * xScale), (int)(e.Location.Y * yScale));                    LineSegment2D ls = new LineSegment2D(p1, p2);                    int thickness = (int)(LineWidth * xScale);                    imageSourceClone.Draw(ls, new Bgr(255d, 255d, 255d), thickness);                    pbSource.Image = imageSourceClone.Bitmap;                    imageMarkers.Draw(ls, new Gray(drawCount), thickness);                }                previousMouseLocation = e.Location;            }        }        //当松开鼠标左键时，将绘图的前一位置设置为（-1，-1）        private void pbSource_MouseUp(object sender, MouseEventArgs e)        {            previousMouseLocation = new Point(-1, -1);            drawCount++;        }



图像分割 Image Segmentation






 


 


        您可以用类似下面的方式来使用分水岭算法：



图像分割 Image Segmentation使用分水岭分割算法


图像分割 Image Segmentation





        /// <summary>        /// 分水岭算法图像分割        /// </summary>        /// <returns>返回用时</returns>        private string Watershed()        {            //分水岭算法分割            Image<Gray, Int32> imageMarkers2 = imageMarkers.Copy();            Stopwatch sw = new Stopwatch();            sw.Start();            CvInvoke.cvWatershed(imageSource.Ptr, imageMarkers2.Ptr);            sw.Stop();            //将分割的结果转换到256级灰度图像            pbResult.Image = imageMarkers2.Bitmap;            imageMarkers2.Dispose();            return string.Format("分水岭图像分割，用时：{0:F05}毫秒。\r\n", sw.Elapsed.TotalMilliseconds);        }



图像分割 Image Segmentation






 


 


金字塔分割算法
    金字塔分割算法由cvPrySegmentation所实现，该函数的使用很简单；需要注意的是图像的尺寸以及金字塔的层数，图像的宽度和高度必须能被2整除，能够被2整除的次数决定了金字塔的最大层数。下面的代码演示了如果校验金字塔层数：



图像分割 Image Segmentation校验金字塔分割的金字塔层数


图像分割 Image Segmentation





        /// <summary>        /// 当改变金字塔分割的参数“金字塔层数”时，对参数进行校验        /// </summary>        /// <param name="sender"></param>        /// <param name="e"></param>        private void txtPSLevel_TextChanged(object sender, EventArgs e)        {            int level = int.Parse(txtPSLevel.Text);            if (level < 1 || imageSource.Width % (int)(Math.Pow(2, level - 1)) != 0 || imageSource.Height % (int)(Math.Pow(2, level - 1)) != 0)                MessageBox.Show(this, "注意：您输入的金字塔层数不符合要求，计算结果可能会无效。", "金字塔层数错误");        }



图像分割 Image Segmentation






 


 


使用金字塔分割的示例代码如下：



图像分割 Image Segmentation使用金字塔分割算法


图像分割 Image Segmentation





        /// <summary>        /// 金字塔分割算法        /// </summary>        /// <returns></returns>        private string PrySegmentation()        {            //准备参数            Image<Bgr, Byte> imageDest = new Image<Bgr, byte>(imageSource.Size);            MemStorage storage = new MemStorage();            IntPtr ptrComp = IntPtr.Zero;            int level = int.Parse(txtPSLevel.Text);            double threshold1 = double.Parse(txtPSThreshold1.Text);            double threshold2 = double.Parse(txtPSThreshold2.Text);            //金字塔分割            Stopwatch sw = new Stopwatch();            sw.Start();            CvInvoke.cvPyrSegmentation(imageSource.Ptr, imageDest.Ptr, storage.Ptr, out ptrComp, level, threshold1, threshold2);            sw.Stop();            //显示结果            pbResult.Image = imageDest.Bitmap;            //释放资源            imageDest.Dispose();            storage.Dispose();            return string.Format("金字塔分割，用时：{0:F05}毫秒。\r\n", sw.Elapsed.TotalMilliseconds);        }



图像分割 Image Segmentation






 


 


均值漂移分割算法
    均值漂移分割算法由cvPryMeanShiftFiltering所实现，均值漂移分割的金字塔层数只能介于[1,7]之间，您可以用类似下面的代码来使用它：



图像分割 Image Segmentation使用均值漂移分割算法


图像分割 Image Segmentation





        /// <summary>        /// 均值漂移分割算法        /// </summary>        /// <returns></returns>        private string PryMeanShiftFiltering()        {            //准备参数            Image<Bgr, Byte> imageDest = new Image<Bgr, byte>(imageSource.Size);            double spatialRadius = double.Parse(txtPMSFSpatialRadius.Text);            double colorRadius = double.Parse(txtPMSFColorRadius.Text);            int maxLevel = int.Parse(txtPMSFNaxLevel.Text);            int maxIter = int.Parse(txtPMSFMaxIter.Text);            double epsilon = double.Parse(txtPMSFEpsilon.Text);            MCvTermCriteria termcrit = new MCvTermCriteria(maxIter, epsilon);            //均值漂移分割            Stopwatch sw = new Stopwatch();            sw.Start();            OpenCvInvoke.cvPyrMeanShiftFiltering(imageSource.Ptr, imageDest.Ptr, spatialRadius, colorRadius, maxLevel, termcrit);            sw.Stop();            //显示结果            pbResult.Image = imageDest.Bitmap;            //释放资源            imageDest.Dispose();            return string.Format("均值漂移分割，用时：{0:F05}毫秒。\r\n", sw.Elapsed.TotalMilliseconds);        }



图像分割 Image Segmentation






 


 


    函数cvPryMeanShiftFiltering在EmguCv中没有实现，我们可以用下面的方式来使用：



图像分割 Image Segmentation调用均值漂移分割





        //均值漂移分割        [DllImport("cv200.dll")]        public static extern void cvPyrMeanShiftFiltering(IntPtr src, IntPtr dst, double spatialRadius, double colorRadius, int max_level, MCvTermCriteria termcrit);







 


 


分割效果及性能对比
    上述三种分割算法的效果如何呢？下面我们以它们的默认参数，对一幅2272x1704大小的图像进行分割。得到的结果如下所示：


图像分割 Image Segmentation


图1 分水岭分割算法（左图白色的线条用于标记区域）


图像分割 Image Segmentation


图2 金字塔分割算法


图像分割 Image Segmentation


图3 均值漂移分割算法
    从上面我们可以看出：
    （1）分水岭分割算法的分割效果效果最好，均值漂移分割算法次之，而金字塔分割算法的效果最差；
    （2）均值漂移分割算法效率最高，分水岭分割算法接近于均值漂移算法，金字塔分割算法需要很长的时间。
    值得注意的是分水岭算法对标记很敏感，需要仔细而认真的绘制。


 


    本文的完整代码如下：



图像分割 Image Segmentation本文完整代码


图像分割 Image Segmentation





using System;using System.Collections.Generic;using System.ComponentModel;using System.Data;using System.Drawing;using System.Linq;using System.Text;using System.Windows.Forms;using System.Diagnostics;using System.Runtime.InteropServices;using Emgu.CV;using Emgu.CV.CvEnum;using Emgu.CV.Structure;using Emgu.CV.UI;namespace ImageProcessLearn{    public partial class FormImageSegment : Form    {        //成员变量        private string sourceImageFileName = "wky_tms_2272x1704.jpg";//源图像文件名        private Image<Bgr, Byte> imageSource = null;                //源图像        private Image<Bgr, Byte> imageSourceClone = null;           //源图像的克隆        private Image<Gray, Int32> imageMarkers = null;              //标记图像        private double xScale = 1d;                                 //原始图像与PictureBox在x轴方向上的缩放        private double yScale = 1d;                                 //原始图像与PictureBox在y轴方向上的缩放        private Point previousMouseLocation = new Point(-1, -1);    //上次绘制线条时，鼠标所处的位置        private const int LineWidth = 5;                            //绘制线条的宽度        private int drawCount = 1;                                  //用户绘制的线条数目，用于指定线条的颜色                public FormImageSegment()        {            InitializeComponent();        }        //窗体加载时        private void FormImageSegment_Load(object sender, EventArgs e)        {            //设置提示            toolTip.SetToolTip(rbWatershed, "可以在源图像上用鼠标绘制大致分割区域线条，该线条用于分水岭算法");            toolTip.SetToolTip(txtPSLevel, "金字塔层数跟图像尺寸有关，该值只能是图像尺寸被2整除的次数，否则将得出错误结果");            toolTip.SetToolTip(txtPSThreshold1, "建立连接的错误阀值");            toolTip.SetToolTip(txtPSThreshold2, "分割簇的错误阀值");            toolTip.SetToolTip(txtPMSFSpatialRadius, "空间窗的半径");            toolTip.SetToolTip(txtPMSFColorRadius, "色彩窗的半径");            toolTip.SetToolTip(btnClearMarkers, "清除绘制在源图像上，用于分水岭算法的大致分割区域线条");            //加载图像            LoadImage();        }        //当窗体关闭时，释放资源        private void FormImageSegment_FormClosing(object sender, FormClosingEventArgs e)        {            if (imageSource != null)                imageSource.Dispose();            if (imageSourceClone != null)                imageSourceClone.Dispose();            if (imageMarkers != null)                imageMarkers.Dispose();        }        //加载源图像        private void btnLoadImage_Click(object sender, EventArgs e)        {            OpenFileDialog ofd = new OpenFileDialog();            ofd.CheckFileExists = true;            ofd.DefaultExt = "jpg";            ofd.Filter = "图片文件|*.jpg;*.png;*.bmp|所有文件|*.*";            if (ofd.ShowDialog(this) == DialogResult.OK)            {                if (ofd.FileName != "")                {                    sourceImageFileName = ofd.FileName;                    LoadImage();                }            }            ofd.Dispose();        }        //清除分割线条        private void btnClearMarkers_Click(object sender, EventArgs e)        {            if (imageSourceClone != null)                imageSourceClone.Dispose();            imageSourceClone = imageSource.Copy();            pbSource.Image = imageSourceClone.Bitmap;            imageMarkers.SetZero();            drawCount = 1;        }        //当鼠标按下并在源图像上移动时，在源图像上绘制分割线条        private void pbSource_MouseMove(object sender, MouseEventArgs e)        {            //如果按下了左键            if (e.Button == MouseButtons.Left)            {                if (previousMouseLocation.X >= 0 && previousMouseLocation.Y >= 0)                {                    Point p1 = new Point((int)(previousMouseLocation.X * xScale), (int)(previousMouseLocation.Y * yScale));                    Point p2 = new Point((int)(e.Location.X * xScale), (int)(e.Location.Y * yScale));                    LineSegment2D ls = new LineSegment2D(p1, p2);                    int thickness = (int)(LineWidth * xScale);                    imageSourceClone.Draw(ls, new Bgr(255d, 255d, 255d), thickness);                    pbSource.Image = imageSourceClone.Bitmap;                    imageMarkers.Draw(ls, new Gray(drawCount), thickness);                }                previousMouseLocation = e.Location;            }        }        //当松开鼠标左键时，将绘图的前一位置设置为（-1，-1）        private void pbSource_MouseUp(object sender, MouseEventArgs e)        {            previousMouseLocation = new Point(-1, -1);            drawCount++;        }        //加载源图像        private void LoadImage()        {            if (imageSource != null)                imageSource.Dispose();            imageSource = new Image<Bgr, byte>(sourceImageFileName);            if (imageSourceClone != null)                imageSourceClone.Dispose();            imageSourceClone = imageSource.Copy();            pbSource.Image = imageSourceClone.Bitmap;            if (imageMarkers != null)                imageMarkers.Dispose();            imageMarkers = new Image<Gray, Int32>(imageSource.Size);            imageMarkers.SetZero();            xScale = 1d * imageSource.Width / pbSource.Width;            yScale = 1d * imageSource.Height / pbSource.Height;            drawCount = 1;        }        //分割图像        private void btnImageSegment_Click(object sender, EventArgs e)        {            if (rbWatershed.Checked)                txtResult.Text += Watershed();            else if (rbPrySegmentation.Checked)                txtResult.Text += PrySegmentation();            else if (rbPryMeanShiftFiltering.Checked)                txtResult.Text += PryMeanShiftFiltering();        }        /// <summary>        /// 分水岭算法图像分割        /// </summary>        /// <returns>返回用时</returns>        private string Watershed()        {            //分水岭算法分割            Image<Gray, Int32> imageMarkers2 = imageMarkers.Copy();            Stopwatch sw = new Stopwatch();            sw.Start();            CvInvoke.cvWatershed(imageSource.Ptr, imageMarkers2.Ptr);            sw.Stop();            //将分割的结果转换到256级灰度图像            pbResult.Image = imageMarkers2.Bitmap;            imageMarkers2.Dispose();            return string.Format("分水岭图像分割，用时：{0:F05}毫秒。\r\n", sw.Elapsed.TotalMilliseconds);        }        /// <summary>        /// 金字塔分割算法        /// </summary>        /// <returns></returns>        private string PrySegmentation()        {            //准备参数            Image<Bgr, Byte> imageDest = new Image<Bgr, byte>(imageSource.Size);            MemStorage storage = new MemStorage();            IntPtr ptrComp = IntPtr.Zero;            int level = int.Parse(txtPSLevel.Text);            double threshold1 = double.Parse(txtPSThreshold1.Text);            double threshold2 = double.Parse(txtPSThreshold2.Text);            //金字塔分割            Stopwatch sw = new Stopwatch();            sw.Start();            CvInvoke.cvPyrSegmentation(imageSource.Ptr, imageDest.Ptr, storage.Ptr, out ptrComp, level, threshold1, threshold2);            sw.Stop();            //显示结果            pbResult.Image = imageDest.Bitmap;            //释放资源            imageDest.Dispose();            storage.Dispose();            return string.Format("金字塔分割，用时：{0:F05}毫秒。\r\n", sw.Elapsed.TotalMilliseconds);        }        /// <summary>        /// 均值漂移分割算法        /// </summary>        /// <returns></returns>        private string PryMeanShiftFiltering()        {            //准备参数            Image<Bgr, Byte> imageDest = new Image<Bgr, byte>(imageSource.Size);            double spatialRadius = double.Parse(txtPMSFSpatialRadius.Text);            double colorRadius = double.Parse(txtPMSFColorRadius.Text);            int maxLevel = int.Parse(txtPMSFNaxLevel.Text);            int maxIter = int.Parse(txtPMSFMaxIter.Text);            double epsilon = double.Parse(txtPMSFEpsilon.Text);            MCvTermCriteria termcrit = new MCvTermCriteria(maxIter, epsilon);            //均值漂移分割            Stopwatch sw = new Stopwatch();            sw.Start();            OpenCvInvoke.cvPyrMeanShiftFiltering(imageSource.Ptr, imageDest.Ptr, spatialRadius, colorRadius, maxLevel, termcrit);            sw.Stop();            //显示结果            pbResult.Image = imageDest.Bitmap;            //释放资源            imageDest.Dispose();            return string.Format("均值漂移分割，用时：{0:F05}毫秒。\r\n", sw.Elapsed.TotalMilliseconds);        }        /// <summary>        /// 当改变金字塔分割的参数“金字塔层数”时，对参数进行校验        /// </summary>        /// <param name="sender"></param>        /// <param name="e"></param>        private void txtPSLevel_TextChanged(object sender, EventArgs e)        {            int level = int.Parse(txtPSLevel.Text);            if (level < 1 || imageSource.Width % (int)(Math.Pow(2, level - 1)) != 0 || imageSource.Height % (int)(Math.Pow(2, level - 1)) != 0)                MessageBox.Show(this, "注意：您输入的金字塔层数不符合要求，计算结果可能会无效。", "金字塔层数错误");        }        /// <summary>        /// 当改变均值漂移分割的参数“金字塔层数”时，对参数进行校验        /// </summary>        /// <param name="sender"></param>        /// <param name="e"></param>        private void txtPMSFNaxLevel_TextChanged(object sender, EventArgs e)        {            int maxLevel = int.Parse(txtPMSFNaxLevel.Text);            if (maxLevel < 0 || maxLevel > 8)                MessageBox.Show(this, "注意：均值漂移分割的金字塔层数只能在0至8之间。", "金字塔层数错误");        }    }}



图像分割 Image Segmentation




            


再分享一下我老师大神的人工智能教程吧。零基础！通俗易懂！风趣幽默！还带黄段子！希望你也加入到我们人工智能的队伍中来！https://blog.csdn.net/jiangjunshow

 
                    
            
                

                  
 
        

	 
        

            

                                            

        

        
相关文章：

        

      
            
 
                    
    
                                                    
  • 
                                 
                                2022-12-23
                            
    • 
                                                        
  • 
                                 
                                2021-06-17
                            
    • 
                                                        
  • 
                                 
                                2022-01-06
                            
    • 
                                                        
  • 
                                 
                                2021-12-13
                            
    • 
                                                        
  • 
                                 
                                2020-02-26
                            
    • 
                                                        
  • 
                                 
                                2022-01-01
                            
    • 
                                                        
  • 
                                 
                                2021-04-12
                            
    • 
                                                

               
   
         

    


  






 
          

    
猜你喜欢 

        
    
                                 
  • 
                     
                    2021-11-04
                
    • 
                                     
  • 
                     
                    2021-12-02
                
    • 
                                     
  • 
                     
                    2021-11-28
                
    • 
                                     
  • 
                     
                    2022-12-23
                
    • 
                                     
  • 
                     
                    2021-04-14
                
    • 
                                     
  • 
                     
                    2021-05-01
                
    • 
                                     
  • 
                     
                    2021-07-27
                
    • 
                            

   
 



    
相关资源 

        
   

   
    

     
相似解决方案 

         
    
                          

    

  
   

    

 
   

    
热门标签  

    

                    Java
                        Python
                        linux
                        javascript
                        Mysql
                        C#
                        Docker
                        算法
                        前端
                        SpringBoot
                        Redis
                        Vue
                        spring
                        设计模式
                        .net core
                        .net
                        kubernetes
                        c++
                        数据库
                        数据结构
                        大数据
                        js
                        机器学习
                        微服务
                        Android
                        Go
                        程序员
                        面试
                        JVM
                        ASP.net core
                        云原生
                        人工智能
                        后端
                        PHP
                        git
                        CSS
                        golang
                        k8s
                        Nginx
                        Django
                        mybatis
                        深度学习
                        多线程
                        React
                        架构
                        devops
                        爬虫
                        云计算
                        Spring Boot
                        LeetCode