Visual Studio C# 试图读取或写入受保护的内存。这通常表明其他内存已损坏

Posted 2023-03-16

【中文标题】Visual Studio C# 试图读取或写入受保护的内存。这通常表明其他内存已损坏

【英文标题】：Visual Studio C# Attempted to read or write protected memory. This is often an indication that other memory is corrupt

【发布时间】：2018-02-19 04:24:21

【问题描述】：

我正在创建一个使用 4 个摄像头进行面部识别的考勤系统。我在 C# 中使用 Emgu CV 3.0。现在，在我的由 4 个图像框组成的考勤表单中，应用程序突然停止并返回主表单，并在引用考勤表单的按钮上显示错误。错误是：

试图读取或写入受保护的内存。这通常表明其他内存已损坏。

这是发生错误的代码：

    private void btn_attendance_Click(object sender, EventArgs e)
    
        attendance attendance = new attendance();
        attendance.ShowDialog();
    

这是不带识别部分的考勤表代码：

public partial class attendance : Form

    private Capture cam1, cam2, cam3, cam4;
    private CascadeClassifier _cascadeClassifier;
    private RecognizerEngine _recognizerEngine;
    private String _trainerDataPath = "\\traineddata_v2";
    private readonly String dbpath = "Server=localhost;Database=faculty_attendance_system;Uid=root;Pwd=root;";
    mysqlConnection conn;

    public attendance()
    
       InitializeComponent();
        conn = new MySqlConnection("Server=localhost;Database=faculty_attendance_system;Uid=root;Pwd=root;");
    

    private void btn_home_Click(object sender, EventArgs e)
    
        this.Close();
    

    private void attendance_Load(object sender, EventArgs e)
    
        time_now.Start();
        lbl_date.Text = DateTime.Now.ToString("");
        _recognizerEngine = new RecognizerEngine(dbpath, _trainerDataPath);

        _cascadeClassifier = new CascadeClassifier(Application.StartupPath + "/haarcascade_frontalface_default.xml");
        cam1 = new Capture(0);
        cam2 = new Capture(1);
        cam3 = new Capture(3);
        cam4 = new Capture(4);

        Application.Idle += new EventHandler(ProcessFrame);
    

    private void ProcessFrame(Object sender, EventArgs args)
    
        Image<Bgr, byte> nextFrame_cam1 = cam1.QueryFrame().ToImage<Bgr, Byte>();
        Image<Bgr, byte> nextFrame_cam2 = cam2.QueryFrame().ToImage<Bgr, Byte>();
        Image<Bgr, byte> nextFrame_cam3 = cam3.QueryFrame().ToImage<Bgr, Byte>();
        Image<Bgr, byte> nextFrame_cam4 = cam4.QueryFrame().ToImage<Bgr, Byte>();

        using (nextFrame_cam1)
        
           if (nextFrame_cam1 != null)
            
                Image<Gray, byte> grayframe = nextFrame_cam1.Convert<Gray, byte>();
                var faces = _cascadeClassifier.DetectMultiScale(grayframe, 1.5, 10, Size.Empty, Size.Empty);
                foreach (var face in faces)
                
                    nextFrame_cam1.Draw(face, new Bgr(Color.Green), 3);
                    var predictedUserId = _recognizerEngine.RecognizeUser(new Image<Gray, byte>(nextFrame_cam1.Bitmap));
                
                imageBox1.Image = nextFrame_cam1;
            
        

        using (nextFrame_cam2)
        
           if (nextFrame_cam2!= null)
            
                Image<Gray, byte> grayframe = nextFrame_cam2.Convert<Gray, byte>();
                var faces = _cascadeClassifier.DetectMultiScale(grayframe, 1.5, 10, Size.Empty, Size.Empty);
                foreach (var face in faces)
                
                    nextFrame_cam2.Draw(face, new Bgr(Color.Green), 3);
                    var predictedUserId = _recognizerEngine.RecognizeUser(new Image<Gray, byte>(nextFrame_cam2.Bitmap));
                
                imageBox2.Image = nextFrame_cam2;
            
        


        using (nextFrame_cam3)
        
           if (nextFrame_cam3!= null)
            
                Image<Gray, byte> grayframe = nextFrame_cam3.Convert<Gray, byte>();
                var faces = _cascadeClassifier.DetectMultiScale(grayframe, 1.5, 10, Size.Empty, Size.Empty);
                foreach (var face in faces)
                
                    nextFrame_cam3.Draw(face, new Bgr(Color.Green), 3);
                    var predictedUserId = _recognizerEngine.RecognizeUser(new Image<Gray, byte>(nextFrame_cam3.Bitmap));
                
                imageBox3.Image = nextFrame_cam3;
            
        

        using (nextFrame_cam4)
        
           if (nextFrame_cam4!= null)
            
                Image<Gray, byte> grayframe = nextFrame_cam4.Convert<Gray, byte>();
                var faces = _cascadeClassifier.DetectMultiScale(grayframe, 1.5, 10, Size.Empty, Size.Empty);
                foreach (var face in faces)
                
                    nextFrame_cam4.Draw(face, new Bgr(Color.Green), 3);
                    var predictedUserId = _recognizerEngine.RecognizeUser(new Image<Gray, byte>(nextFrame_cam4.Bitmap));
                
                imageBox4.Image = nextFrame_cam4;
            
        
    

【问题讨论】：

所以使用调试器并单步执行代码来找出问题的实际位置。我们不能为您做到这一点；我们没有您所有的代码和项目文件和参考资料。

【参考方案1】：

请阅读这篇文章以了解什么是内存泄漏。 http://www.dotnetfunda.com/articles/show/625/best-practices-no-5-detecting-net-application-memory-leaks

您的错误表明您正在创建一个类的许多实例或任何函数的递归调用。 使用 Using() 创建 Emgu 的对象，以便在您的代码终止后立即释放托管或非托管内存。

public partial class attendance : Form

    private Capture cam1, cam2, cam3, cam4;
    private CascadeClassifier _cascadeClassifier;
    private RecognizerEngine _recognizerEngine;
    private String _trainerDataPath = "\\traineddata_v2";
    private readonly String dbpath = "Server=localhost;Database=faculty_attendance_system;Uid=root;Pwd=root;";
    MySqlConnection conn;

    public attendance()
    
        InitializeComponent();
        conn = new MySqlConnection("Server=localhost;Database=faculty_attendance_system;Uid=root;Pwd=root;");
    

    private void btn_home_Click(object sender, EventArgs e)
    
        this.Close();
    
    private void attendance_Load(object sender, EventArgs e)
    
        time_now.Start();
        lbl_date.Text = DateTime.Now.ToString("");
        _recognizerEngine = new RecognizerEngine(dbpath, _trainerDataPath);
        _cascadeClassifier = new CascadeClassifier(Application.StartupPath + "/haarcascade_frontalface_default.xml");
        cam1 = new Capture(0);
        cam2 = new Capture(1);
        cam3 = new Capture(3);
        cam4 = new Capture(4);
        Application.Idle += new EventHandler(ProcessFrame);
    
    private void ProcessFrame(Object sender, EventArgs args)
    
        using (Image<Bgr, byte> nextFrame_cam1 = cam1.QueryFrame().ToImage<Bgr, Byte>())
        
            if (nextFrame_cam1 != null)
            
                Image<Gray, byte> grayframe = nextFrame_cam1.Convert<Gray, byte>();
                var faces = _cascadeClassifier.DetectMultiScale(grayframe, 1.5, 10, Size.Empty, Size.Empty);
                foreach (var face in faces)
                
                    nextFrame_cam1.Draw(face, new Bgr(Color.Green), 3);
                    var predictedUserId = _recognizerEngine.RecognizeUser(new Image<Gray, byte>(nextFrame_cam1.Bitmap));
                
                imageBox1.Image = nextFrame_cam1;
            
        

        using (Image<Bgr, byte> nextFrame_cam2 = cam2.QueryFrame().ToImage<Bgr, Byte>())
        
            if (nextFrame_cam2 != null)
            
                Image<Gray, byte> grayframe = nextFrame_cam2.Convert<Gray, byte>();
                var faces = _cascadeClassifier.DetectMultiScale(grayframe, 1.5, 10, Size.Empty, Size.Empty);
                foreach (var face in faces)
                
                    nextFrame_cam2.Draw(face, new Bgr(Color.Green), 3);
                    var predictedUserId = _recognizerEngine.RecognizeUser(new Image<Gray, byte>(nextFrame_cam2.Bitmap));
                
                imageBox2.Image = nextFrame_cam2;
            
        


        using (Image<Bgr, byte> nextFrame_cam3 = cam3.QueryFrame().ToImage<Bgr, Byte>())
        
            if (nextFrame_cam3 != null)
            
                Image<Gray, byte> grayframe = nextFrame_cam3.Convert<Gray, byte>();
                var faces = _cascadeClassifier.DetectMultiScale(grayframe, 1.5, 10, Size.Empty, Size.Empty);
                foreach (var face in faces)
                
                    nextFrame_cam3.Draw(face, new Bgr(Color.Green), 3);
                    var predictedUserId = _recognizerEngine.RecognizeUser(new Image<Gray, byte>(nextFrame_cam3.Bitmap));
                
                imageBox3.Image = nextFrame_cam3;
            
        

        using (Image<Bgr, byte> nextFrame_cam4 = cam4.QueryFrame().ToImage<Bgr, Byte>())
        
            if (nextFrame_cam4 != null)
            
                Image<Gray, byte> grayframe = nextFrame_cam4.Convert<Gray, byte>();
                var faces = _cascadeClassifier.DetectMultiScale(grayframe, 1.5, 10, Size.Empty, Size.Empty);
                foreach (var face in faces)
                
                    nextFrame_cam4.Draw(face, new Bgr(Color.Green), 3);
                    var predictedUserId = _recognizerEngine.RecognizeUser(new Image<Gray, byte>(nextFrame_cam4.Bitmap));
                
                imageBox4.Image = nextFrame_cam4;
            
        
    

请按照本文档了解使用 EMGU.CV 进行人脸识别的标准方法。 http://www.emgu.com/wiki/index.php/Face_detection

【讨论】：

欢迎来到 SO。如果可能，最好引用链接中最重要的部分（同时保留链接作为参考）。这样你的答案在未来几年就不会过时。谢谢

感谢@LonelyNeuron