Develop Reference

c# color extractor frequently returning white

I have a program that has a small graphics portion to it, where you can look at and store the color of specific pixels of the screen.
The problem is that when I extract color, sometimes I get the color I want... sometimes I just get White, or some other random color (usually white, but I'm also using a white background to test with).
Can you spot any reason why that would be? I'm also totally open to possible improvements.
I have a picturebox called Canvas, and I can either click "Capture Image" which does a screenshot as follows:
private void CaptureImage_Click(object sender, EventArgs e)
{
this.Visible = false;
Bitmap bitmap = new Bitmap(Screen.PrimaryScreen.Bounds.Width, Screen.PrimaryScreen.Bounds.Height);
Graphics graphics = Graphics.FromImage(bitmap as Image);
graphics.CopyFromScreen(0, 0, 0, 0, bitmap.Size);
Canvas.SizeMode = PictureBoxSizeMode.AutoSize;
Canvas.Image = bitmap;
graphics.Dispose();
this.Visible = true;
}
Or I can load an image file, though for this test I'm just doing the capture.
When I right-click somewhere on the image, in my mousedown event I do:
if (e.Button == MouseButtons.Right)
{
pointPoint = e.Location;
Debug.WriteLine("right mouse: " + e.Location.ToString());
ContextMenuStrip cm = new ContextMenuStrip();
cm.Items.Add("Flag this as Point");
cm.Items.Add("Extract Color");
Canvas.ContextMenuStrip = cm;
cm.ItemClicked += new ToolStripItemClickedEventHandler(Cm2Click);
}
(pointPoint is a Point object)
As you can see, I verify the location where I clicked using the debug window. In my context menu, I also verify it again to make sure it didn't change before getting called. The context menu does:
private void Cm2Click(object sender, ToolStripItemClickedEventArgs e)
{
//var clickedMenuItem = sender as MenuItem;
ToolStripItem menuText = e.ClickedItem;
Debug.WriteLine(menuText);
if (menuText.ToString() == "Extract Color")
{
string input = "Default";
bool testBox = false;
do
{
if (InputBox.inputBox("ColorName", "You must provide a unique name for this color", ref input) == DialogResult.OK)
{
if (!String.IsNullOrWhiteSpace(input) && input != "Default")
{
Debug.WriteLine("location changed to: " + pointPoint);
colorExtractor = Functions.GetColorAt(pointPoint);
ColorData clr = new ColorData(input, colorExtractor);
string expression;
expression = $"Name = '{clr.Name}'";
DataRow[] foundRows;
DataTable dt = _allEventData.Tables["ColorData"];
foundRows = dt.Select(expression);
if (foundRows.Length > 0) {
testBox = false;
Debug.WriteLine("Color Found");
}
else {
dt.Rows.Add(clr.Name, clr.R, clr.G, clr.B);
dt.AcceptChanges();
testBox = true;
}
}
} else
{
return;
}
} while (testBox == false);
}
}
and ColorData is just a class I defined to help store data:
using System;
using System.Collections.Generic;
using System.Linq;
using System.Text;
using System.Threading.Tasks;
using System.Drawing;
namespace EventClicker
{
public class ColorData : EventArgs
{
public string Name { get; set; }
public int R { get; set; }
public int G { get; set; }
public int B { get; set; }
public bool Found { get; set; }
public ColorData(String name, Color color)
{
this.Name = name;
this.R = color.R;
this.G = color.G;
this.B = color.B;
}
public ColorData(String name, String r, String g, String b)
{
this.Name = name;
this.R = Int32.Parse(r);
this.G = Int32.Parse(g);
this.B = Int32.Parse(b);
}
}
}
Last but not least, the method I call (Functions.GetColorAt) is defined here (and I definitely stole this from someone on the web) Edit: That somewhere is here, How to read the Color of a Screen Pixel
using System;
using System.Collections.Generic;
using System.Drawing;
using System.Drawing.Imaging;
using System.Linq;
using System.Runtime.InteropServices;
using System.Text;
using System.Threading.Tasks;
namespace EventClicker
{
public static class Functions
{
private static Bitmap screenPixel = new Bitmap(1, 1, PixelFormat.Format32bppArgb);
[DllImport("gdi32.dll", CharSet = CharSet.Auto, SetLastError = true, ExactSpelling = true)]
public static extern int BitBlt(IntPtr hDC, int x, int y, int nWidth, int nHeight, IntPtr hSrcDC, int xSrc, int ySrc, int dwRop);
public static Color GetColorAt(Point location)
{
using (Graphics gdest = Graphics.FromImage(screenPixel))
{
using (Graphics gsrc = Graphics.FromHwnd(IntPtr.Zero))
{
IntPtr hSrcDC = gsrc.GetHdc();
IntPtr hDC = gdest.GetHdc();
int retval = BitBlt(hDC, 0, 0, 1, 1, hSrcDC, location.X, location.Y, (int)CopyPixelOperation.SourceCopy);
gdest.ReleaseHdc();
gsrc.ReleaseHdc();
}
}
return screenPixel.GetPixel(0, 0);
}
}
}

GDI+ Game Engine (What can I do to make my engine run faster?) [duplicate]

This question already exists:
How to make my game engine faster (GDI+, C#)
Closed 6 years ago.
I'm currently working on a 2D Game Engine using GDI+. Everything is running quite smoothly, however I know there must be a few more things I can do to make my engine render graphics faster.
At the moment if I rendered an empty bitmap that's 800x600 pixels to the screen, I get about 130fps. But once I draw an image that's 800x600 pixels on the bitmap, I get about 70fps. That's not an issue at all, I'm actually proud of the results, I just know it can be better.
I was just wondering if anyone had any suggestions as to what might make the engine run even faster?
Here's the code to my window class:
using System;
using System.Windows.Forms;
using GameEngine.Graphics;
using System.Threading;
using System.Drawing;
namespace GameEngine.Core
{
public class Game : Form
{
// Just a class that checks the fps and such
private GameTime gameTime;
private GraphicsEngine graphicsEngine;
private bool isLoaded = false;
private bool isRunning = false;
public Game(int width, int height, string title)
{
this.Width = width;
this.Height = height;
this.Text = title;
this.MaximizeBox = false;
this.FormBorderStyle = FormBorderStyle.FixedSingle;
this.SetStyle(ControlStyles.AllPaintingInWmPaint, true);
this.SetStyle(ControlStyles.OptimizedDoubleBuffer, true);
this.Paint += Game_Paint;
this.FormClosing += Game_FormClosing;
gameTime = new GameTime();
graphicsEngine = new GraphicsEngine(this);
Show();
}
public void Run()
{
Initialize();
Load();
isRunning = true;
while (isRunning && isLoaded)
{
// Yes I know, this is frowned upon
Application.DoEvents();
gameTime.Update();
// Only update if the form is focused and if the GameTime class says we're allowed
if (this.Focused && gameTime.Cycled)
{
BeginUpdate();
UpdateGame();
LateUpdate();
Render();
}
else
{
Thread.Sleep(1);
}
}
Exit();
}
public void Exit()
{
Unload();
this.Close();
// Using console application, so lets exit the console
Environment.Exit(0);
}
private void Initialize()
{
gameTime.Initialize();
}
private new void Load()
{
isLoaded = true;
}
private void Unload()
{
}
private void BeginUpdate()
{
}
private void UpdateGame()
{
this.Title = "FPS: " + gameTime.FPS;
}
private void LateUpdate()
{
}
Bitmap bmp = new Bitmap("Test.jpg");
private void Render()
{
// Draw the nice tree to the window
graphicsEngine.DrawBitmap(bmp, 0, 0);
// Cause the window to redraw it's self
this.Invalidate();
}
private void Game_Paint(object sender, PaintEventArgs e)
{
// Lets just make sure once again that we're allowed to draw
if (gameTime.Cycled)
{
// Render the graphics to the screen
graphicsEngine.Render(e.Graphics);
}
}
private void Game_FormClosing(object sender, FormClosingEventArgs e)
{
isLoaded = false;
}
public string Title
{
get { return Text; }
set { Text = value; }
}
}
}
and here's the graphics engine class:
using System.Drawing;
using GameEngine.Core;
using System.Drawing.Drawing2D;
using System.Drawing.Text;
namespace GameEngine.Graphics
{
public class GraphicsEngine
{
private Game game;
private System.Drawing.Graphics backBuffer;
private System.Drawing.Bitmap backBufferBitmap;
public GraphicsEngine(Game game)
{
this.game = game;
backBufferBitmap = new Bitmap(800, 600);
backBuffer = System.Drawing.Graphics.FromImage(backBufferBitmap);
backBuffer.CompositingMode = CompositingMode.SourceOver;
backBuffer.CompositingQuality = CompositingQuality.HighSpeed;
backBuffer.SmoothingMode = SmoothingMode.None;
backBuffer.InterpolationMode = InterpolationMode.Low;
backBuffer.TextRenderingHint = TextRenderingHint.SystemDefault;
backBuffer.PixelOffsetMode = PixelOffsetMode.Half;
}
public void Render(System.Drawing.Graphics g)
{
g.CompositingMode = CompositingMode.SourceCopy;
g.CompositingQuality = CompositingQuality.AssumeLinear;
g.SmoothingMode = SmoothingMode.None;
g.InterpolationMode = InterpolationMode.NearestNeighbor;
g.TextRenderingHint = TextRenderingHint.SystemDefault;
g.PixelOffsetMode = PixelOffsetMode.HighSpeed;
g.DrawImage(backBufferBitmap, new Rectangle(0, 0, game.Width, game.Height), new Rectangle(0, 0, game.Width, game.Height), GraphicsUnit.Pixel);
backBuffer.Clear(Color.Black);
}
public void DrawString(string text, int x, int y)
{
backBuffer.DrawString(text, SystemFonts.DefaultFont, Brushes.White, x, y);
}
public void DrawBitmap(Bitmap bitmap, int x, int y)
{
backBuffer.DrawImage(bitmap, x, y);
}
public void DrawBitmap(Bitmap bitmap, int x, int y, int width, int height)
{
backBuffer.DrawImageUnscaled(bitmap, x, y, width, height);
}
}
}

You should try to operate a kind of Logical thread separated from your drawing thread. Maybe replace the thread sleeping in your logical by a (tick management).
As asked in comments below, the fact to separate UI & logic in different thread could help you to have a better flow in both tasks. The hard part is to keep UI and logic Synced, but you're less exposed to slow methods/function which would slow down UI.

Deleting JPEG Not Allowed(Used By Another Process)

Question Edited!! This is an example of compilabile files to help you find my problem!
I save some screenshots in a specific file.
When i use function DeleteFile() it should delete first the contents and then the folder but then i get "that icon.jpg is used by another process"!
(Either way to delete a folder it has to have no content!)
ImageExtensions.cs
using System.Drawing;
using System.Drawing.Imaging;
using System.IO;
namespace name
{
static class ImageExtensions
{
public static void SaveToFile(Image image,string path,ImageFormat format)
{
using (var stream = File.OpenWrite(path))
{
image.Save(stream, format);
}
}
}
}
ScreenCapture.cs
using System;
using System.Drawing;
using System.Runtime.InteropServices;
namespace name
{
public class ScreenCapture
{
[DllImport("user32.dll")]
private static extern IntPtr GetForegroundWindow();
[DllImport("user32.dll")]
private static extern IntPtr GetWindowRect(IntPtr hWnd, ref Rect rect);
[StructLayout(LayoutKind.Sequential)]
private struct Rect
{
public int Left;
public int Top;
public int Right;
public int Bottom;
}
[DllImport("user32.dll", CharSet = CharSet.Auto, ExactSpelling = true)]
public static extern IntPtr GetDesktopWindow();
public static Image CaptureDesktop()
{
return CaptureWindow(GetDesktopWindow());
}
public static Bitmap CaptureActiveWindow()
{
return CaptureWindow(GetForegroundWindow());
}
public static Bitmap CaptureWindow(IntPtr handle)
{
var rect = new Rect();
GetWindowRect(handle, ref rect);
var bounds = new Rectangle(rect.Left, rect.Top, rect.Right - rect.Left, rect.Bottom - rect.Top);
var result = new Bitmap(bounds.Width, bounds.Height);
using (var graphics = Graphics.FromImage(result))
{
graphics.CopyFromScreen(new Point(bounds.Left, bounds.Top), Point.Empty, bounds.Size);
}
return result;
}
}
}
AnotherFile.cs (Uses PrintScreen() )
public void PrintScreen()
{
using (var image = ScreenCapture.CaptureDesktop())
{
ImageExtensions.SaveToFile(image, (file_images + ".jpg"), ImageFormat.Jpeg);
}
}
Form1.cs (Main)
using ...
using ...
namespace name{
public partial class Form1 : Form
{
const int MaxRetries = 3;
const int DelayBetweenRetries = 1000;
const int ERROR_SHARING_VIOLATION = unchecked((int)0x80070020);
string file_images = my_path_of_images_file;
Public Form1()
{
InitializeComponent();
CreateFile();
}
private void Form1_Load(object sender, EventArgs e){}
public void CreateFile()
{
if (!Directory.Exists(file_images))
{
DirectoryInfo di = Directory.CreateDirectory(file_images);
}
}
public void DeleteFiles()
{
string[] filesToDelete = Directory.GetFiles(file_images);
if (filesToDelete != null)
{
var files = filesToDelete.Where(x => Path.GetExtension(x).Contains(".jpg"));
foreach (var file in files)
{
var temp = file;
DeleteFile(temp); // Delete JPEG
}
}
Directory.Delete(file_images); // Delete Folder
}
public static void DeleteFile(string path)
{
for (int i = 0; i < MaxRetries; ++i)
{
try
{
File.Delete(path);
}
catch (IOException e)
{
// You may also sleep whatever the error is...
if (Marshal.GetHRForException(e) == ERROR_SHARING_VIOLATION)
{
Thread.Sleep(DelayBetweenRetries);
continue;
}
throw;
}
}
}
} // End Of Main.cs
} // End Of Namespace

I have had this problem before and I solved it by copying the image data from the filestream into a memorystream, then loading the image from it. Similarly with the saving the image. This way, even if Image keeps a reference to the source or saved stream, it is not a physical file.
A more complete solution: https://siderite.dev/blog/imagefromstream-or-imagefromfile.html
try:
using (var ms=new MemoryStream()) {
image.Save(ms,format);
File.WriteAllBytes(path,ms.ToArray());
}
I wrote this inline, so tweak it if it doesn't work exactly.

Cursor doesn't hide while taking a screenshot

Simple question. I want to take a screenshot of a part of my screen without the mouse pointer in it. I tried this but it doesn't work.
private void button1_Click(object sender, EventArgs e)
{
Cursor.Hide();
gfxScreenshot.CopyFromScreen(//...);
Cursor.Show();
}
The code above was on a button_click event. I transferred the code in a timer_tick event except from the Cursor.Hide() and made the timer's interval 1000. The timer starts when the button is clicked.
private void button1_Click(object sender, EventArgs e)
{
Cursor.Hide();
timer1.Start();
}
private void timer1_Tick(object sender, EventArgs e)
{
gfxScreenshot.CopyFromScreen(//...);
Cursor.Show();
timer1.Stop();
}
It works this way but I have to wait 1 sec. When I reduce the interval to 100 the pointer is visible on the image.
I can only assume that the hide method is slower than the CopyFromScreen method...
Is there any way to make it work without the 1 sec delay??

Get cursor location, move to (0,0), take screenshot, put back cursor. The code works using APIs:
using System.Drawing;
using System.Runtime.InteropServices;
using System.Windows.Forms;
namespace Test1
{
public partial class Form1 : Form
{
[DllImport("user32.dll")]
public static extern bool SetCursorPos(int X, int Y);
[DllImport("user32.dll")]
public static extern bool GetCursorPos(out POINT lpPoint);
[StructLayout(LayoutKind.Sequential)]
public struct POINT
{
public int X;
public int Y;
public static implicit operator Point(POINT point)
{
return new Point(point.X, point.Y);
}
}
public Form1()
{
InitializeComponent();
POINT lpPoint;
//Get current location of cursor
GetCursorPos( out lpPoint );
//Move to (0,0)
SetCursorPos( 0, 0 );
//Take screenshot
//gfxScreenshot.CopyFromScreen(//...);
MessageBox.Show("just for create a delay", "Debug", MessageBoxButtons.OK, MessageBoxIcon.Information);
//Put back cursor
SetCursorPos( lpPoint.X, lpPoint.Y );
}
}
}

WinForms: Detect when cursor enters/leaves the form or its controls

I need a way of detecting when the cursor enters or leaves the form. Form.MouseEnter/MouseLeave doesn't work when controls fill the form, so I will also have to subscribe to MouseEnter event of the controls (e.g. panels on the form). Any other way of tracking form cursor entry/exit globally?

You can try this :
private void Form3_Load(object sender, EventArgs e)
{
MouseDetector m = new MouseDetector();
m.MouseMove += new MouseDetector.MouseMoveDLG(m_MouseMove);
}
void m_MouseMove(object sender, Point p)
{
Point pt = this.PointToClient(p);
this.Text = (this.ClientSize.Width >= pt.X &&
this.ClientSize.Height >= pt.Y &&
pt.X > 0 && pt.Y > 0)?"In":"Out";
}
The MouseDetector class :
using System;
using System.Collections.Generic;
using System.Linq;
using System.Text;
using System.Runtime.InteropServices;
using System.Windows.Forms;
using System.Drawing;
class MouseDetector
{
#region APIs
[DllImport("gdi32")]
public static extern uint GetPixel(IntPtr hDC, int XPos, int YPos);
[DllImport("user32.dll", CharSet = CharSet.Auto)]
public static extern bool GetCursorPos(out POINT pt);
[DllImport("User32.dll", CharSet = CharSet.Auto)]
public static extern IntPtr GetWindowDC(IntPtr hWnd);
#endregion
Timer tm = new Timer() {Interval = 10};
public delegate void MouseMoveDLG(object sender, Point p);
public event MouseMoveDLG MouseMove;
public MouseDetector()
{
tm.Tick += new EventHandler(tm_Tick); tm.Start();
}
void tm_Tick(object sender, EventArgs e)
{
POINT p;
GetCursorPos(out p);
if (MouseMove != null) MouseMove(this, new Point(p.X,p.Y));
}
[StructLayout(LayoutKind.Sequential)]
public struct POINT
{
public int X;
public int Y;
public POINT(int x, int y)
{
X = x;
Y = y;
}
}
}

You can do it with win32 like in this answer:
How to detect if the mouse is inside the whole form and child controls in C#?
Or you could just hook up all the top level controls in OnLoad of the form:
foreach (Control control in this.Controls)
control.MouseEnter += new EventHandler(form_MouseEnter);