We Keep Coding

Image is not drawn at the correct spot

Bitmap image = ReadBitmap("image.png");
Bitmap imageCopy = new Bitmap(image);
Bitmap canvas = new Bitmap(imageCopy.Width+100, imageCopy.Height);
// From this bitmap, the graphics can be obtained, because it has the right PixelFormat
using(Graphics g = Graphics.FromImage(canvas))
{
// Draw the original bitmap onto the graphics of the new bitmap
g.DrawImage(image, 0, 0);
}
// Use tempBitmap as you would have used originalBmp
InputPictureBox.Image = image;
OutputPictureBox.Image = canvas;
I haven't understood the output of this c# code.
The original image is not placed at the correct position. It should have been at (0, 0).
Also, I need a black background.
So, what is going on and how to correct this?

You are loading an Image, then a copy of this source is created using:
Bitmap bitmap = new Bitmap();
When you create a copy of an Image this way, you sacrifice/alter some details:
Dpi Resolution: if not otherwise specified, the resolution is set to the UI resolution. 96 Dpi, as a standard; it might be different with different screen resolutions and scaling. The System in use also affects this value (Windows 7 and Windows 10 will probably/possibly provide different values)
PixelFormat: If not directly copied from the Image source or explicitly specified, the PixelFormat is set to PixelFormat.Format32bppArgb.
From what you were saying, you probably wanted something like this:
var imageSource = Image.FromStream(new MemoryStream(File.ReadAllBytes(#"[SomeImageOfLena]"))), true, false)
var imageCopy = new Bitmap(imageSource.Width + 100, imageSource.Height, imageSource.PixelFormat))
imageCopy.SetResolution(imageSource.HorizontalResolution, imageSource.VerticalResolution);
using (var g = Graphics.FromImage(imageCopy)) {
g.Clear(Color.Black);
g.CompositingMode = CompositingMode.SourceCopy;
g.InterpolationMode = InterpolationMode.HighQualityBicubic;
g.DrawImage(imageSource, (imageCopy.Width - imageSource.Width) / 2, 0);
pictureBox1.Image?.Dispose();
pictureBox2.Image?.Dispose();
pictureBox1.Image = imageSource;
pictureBox2.Image = imageCopy;
}
This is the result:
(The upper/lower frame black color is actually the Picturebox background color)
When the original Image Dpi Resolution is different from the base Dpi Resolution used when creating an Image copy with new Bitmap(), your results may be different from what is expected.
This is what happens with a source Image of 150, 96 and 72 Dpi in the same scenario:
Another important detail is the IDisposable nature of the Image object.
When you create one, you have to Dispose() of it; explicitly, calling the Dispose method, or implicitly, enclosing the Image contructor in a Using statement.
Also, possibly, don't assign an Image object directly loaded from a FileStream.
GDI+ will lock the file, and you will not be able to copy, move or delete it.
With the file, all resources tied to the Images will also be locked.
Make a copy with new Bitmap() (if you don't care of the above mentioned details), or with Image.Clone(), which will preserve the Image Dpi Resolution and PixelFormat.

I am not completely clear on what you are actually needing to do. But anyway, here is a WPF-friendly example of how to draw an image at a specific position inside another image.
Note if all you want to do is display the image in different size and/or put a black border around it, there are much simpler ways to do simply that, without having to create a second image, such as just laying out the image inside a panel that already has the border style you want.
Notice that I am using classes from the System.Windows.Media namespace because that is what WPF uses. These don't mix easily with the older classes from System.Drawing namespace (some of the class names conflict, and Microsoft's .Net framework lacks built-in methods for converting objects between those types), so normally one needs to simply decide whether to use one or the other sets of drawing tools. I assume you have been trying to use System.Drawing. Each has its own pros and cons that would take too long to explain here.
// using System.Windows.Media;
// using System.Windows.Media.Imaging;
private void DrawTwoImages()
{
// For InputPictureBox
var file = new Uri("C:\\image.png");
var inputImage = new BitmapImage(file);
// If your image is stored in a Resource Dictionary, instead use:
// var inputImage = (BitmapImage) Resources["image.png"];
InputPicture.Source = inputImage;
// imageCopy isn't actually needed for this example.
// But since you had it in yours, here is how it's done, anyway.
var imageCopy = inputImage.Clone();
// Parameters for setting up our output picture
int leftMargin = 50;
int topMargin = 5;
int rightMargin = 50;
int bottomMargin = 5;
int width = inputImage.PixelWidth + leftMargin + rightMargin;
int height = inputImage.PixelHeight + topMargin + bottomMargin;
var backgroundColor = Brushes.Black;
var borderColor = (Pen) null;
// Use a DrawingVisual and DrawingContext for drawing
DrawingVisual dv = new DrawingVisual();
using (DrawingContext dc = dv.RenderOpen())
{
// Draw the black background
dc.DrawRectangle(backgroundColor, borderColor, new Rect(0, 0, width, height));
// Copy input image onto output image at desired position
dc.DrawImage(inputImage, new Rect(leftMargin, topMargin,
inputImage.PixelWidth, inputImage.PixelHeight));
}
// For displaying output image
var rtb = new RenderTargetBitmap( width, height, 96, 96, PixelFormats.Pbgra32 );
rtb.Render(dv);
OutputPicture.Source = rtb;
}

How to properly cope with Memory residing Bitmap resize

I have bitmap residing in memory (coming from my webcam but I don't think that this makes a difference.
It is 960x540 120dpi
you see that the picture in the lower part gets till the point where my shirt begins.
I know the bmp dimensions since I put this code prior resize
using (var fileStream = new FileStream(#"C:\temp\3.bmp", FileMode.Create))
{
BitmapEncoder encoder = new PngBitmapEncoder();
encoder.Frames.Add(BitmapFrame.Create(source));
encoder.Save(fileStream);
}
and the result is the picture above.
Then I resize it with that:
var resizedImage = new RenderTargetBitmap(
width, height, // Resized dimensions 200x112
source.DpiX, source.DpiY,// Default DPI values
PixelFormats.Default); // Default pixel format
and the result is the bmp below which is properly 200x112 but it cuts out in part of the image in the lower and right part.
I have seen that the problem is related with the dpi value in the RenderTargetBitmap instruction. If I divide the dpi by 1.25 everything gets fine but why 1.25???????
Thank you in advance for any help
Patrick
--ADD--
There is something additional that I can't understand: I know the initial bitmap size for I have saved it to filesytem with the instructions above.
But if I look at the properties by putting a breakpoint I see:
BITMAP BEFORE RESIZE
width,height = 768, 432
pixelwidth, pixelheight = 960, 540
dpiX, dpiY= 120, 120
BITMAP AFTER RESIZE
width,height = 160, 89
pixelwidth, pixelheight = 200, 112
dpiX, dpiY= 120, 120
now I know that what counts here is pixelwidth, pixelheight so that's correct.
If I do 960/786 I get 1.25! So that's my number but why? Can I correct the code as to make it a general solution???

Instead of a RenderTargetBitmap you should simply use a TransformedBitmap with an appropriate ScaleTransform:
var scale = 200d / 960d;
var resizedImage = new TransformedBitmap(source, new ScaleTransform(scale, scale));

The fastest way to convert canvas to the writeablebitmap in WPF?

I currently have one writeablebitmap image and canvas with drawings and I want to send the images to the peer.In order to reduce the bandwidth, I would like to convert canvas to the writeablebitmap, thus I can blit both the images to a new writeablebitmap. The problem is I cannot find a good way to convert the canvas.
Therefore, I would like to ask if there is a direct way to convert the canvas to a writeablebitmap class.

This is taken from this blog post but instead of writing to a file, it writes to a WriteableBitmap.
public WriteableBitmap SaveAsWriteableBitmap(Canvas surface)
{
if (surface == null) return null;
// Save current canvas transform
Transform transform = surface.LayoutTransform;
// reset current transform (in case it is scaled or rotated)
surface.LayoutTransform = null;
// Get the size of canvas
Size size = new Size(surface.ActualWidth, surface.ActualHeight);
// Measure and arrange the surface
// VERY IMPORTANT
surface.Measure(size);
surface.Arrange(new Rect(size));
// Create a render bitmap and push the surface to it
RenderTargetBitmap renderBitmap = new RenderTargetBitmap(
(int)size.Width,
(int)size.Height,
96d,
96d,
PixelFormats.Pbgra32);
renderBitmap.Render(surface);
//Restore previously saved layout
surface.LayoutTransform = transform;
//create and return a new WriteableBitmap using the RenderTargetBitmap
return new WriteableBitmap(renderBitmap);
}

TransformedBitmap Scaling Mode

I'm using a TransformedBitmap class to draw scaled images to a Bitmap using TransformedBitmap.CopyPixels. Is there a way to specify the scaling mode that is used? RenderOptions.SetBitmapScalingMode doesn't seem to affect anything. I would like to use nearest neighbor but it appears to use a bi linear filter of some sort.

It is not possible to specify the scaling algorithm, it is by design.
The RenderOptions.SetBitmapScalingMode applies to rendering only, e.g. you have a 32*32 icon and want to show it at 256*256 but still in a blocky way (nearest neighbour)
Update
A few ways on how you could overcome this issue :
Do it by yourself :
http://tech-algorithm.com/articles/nearest-neighbor-image-scaling/
Using Forms :
https://stackoverflow.com/a/1856362/361899
Custom drawing :
How to specify the image scaling algorithm used by a WPF Image?
There is AForge too but that might be overkill for your needs.
Update 2
WriteableBitmapEx will probably do the job easily for you : http://writeablebitmapex.codeplex.com/
You can resize a WriteableBitmap, specify interpolation mode and there is nearest neighbor.
Both TransformedBitmap and WriteableBitmapEx inherit from BitmapSource, likely you'll have no change to make at all to your existing code.

public static class Extensions
{
public static BitmapFrame Resize(this
BitmapSource photo, int width, int height,
BitmapScalingMode scalingMode)
{
var group = new DrawingGroup();
RenderOptions.SetBitmapScalingMode(
group, scalingMode);
group.Children.Add(
new ImageDrawing(photo,
new Rect(0, 0, width, height)));
var targetVisual = new DrawingVisual();
var targetContext = targetVisual.RenderOpen();
targetContext.DrawDrawing(group);
var target = new RenderTargetBitmap(
width, height, 96, 96, PixelFormats.Default);
targetContext.Close();
target.Render(targetVisual);
var targetFrame = BitmapFrame.Create(target);
return targetFrame;
}
}
Took from http://weblogs.asp.net/bleroy/resizing-images-from-the-server-using-wpf-wic-instead-of-gdi

How to merge layers and save as an image in WPF?

I have two layers. The first layer is Image control. The source of it is Bitmap Image. And this is the background layer. The second one, which is the front layer is a canvas on which I can draw geometry objects (such as line, polyline, rectangle etc.) and the background of canvas is transparent.
How can I merge these two layers and save it as an image using WPF.

What do you mean by "Layers"? Just two controls sat in the same cell of a grid? If you have both "layers" sat in another container (such as a grid, or even the window) then you can use RenderTargetBitmap with that container to get your image. I have some details, an extension method for taking WPF "Screenshots" on my blog.

You can get bitmap of your parent panel on which you have placed your image control and canvas.
How is the code to get the bitmap of UIElement in WPF.
RenderTargetBitmap bmp = new RenderTargetBitmap(Width, Height, 96, 96, PixelFormats.Pbgra32);
bmp.Render(parentPanel);

Use something like, call this method from your canvas (this is canvas) -
private Bitmap ImageGenerator()
{
var transform = this.LayoutTransform;
// Call UpdateLayout to make sure changes all changes
// while drawing objects on canvas are reflected
var layer = AdornerLayer.GetAdornerLayer(this);
layer?.UpdateLayout();
// Get the size of canvas
var size = new System.Windows.Size(this.ActualWidth, this.ActualHeight);
// Measure and arrange the surface
// VERY IMPORTANT
this.Measure(size);
this.Arrange(new Rect(RenderSize));
RenderTargetBitmap renderBitmap =
new RenderTargetBitmap(
(int)this.ActualWidth,
(int)this.ActualHeight,
96d,
96d,
PixelFormats.Pbgra32);
renderBitmap.Render(this);
JpegBitmapEncoder encoder = new JpegBitmapEncoder();
// push the rendered bitmap to it
encoder.Frames.Add(BitmapFrame.Create(renderBitmap));
var stream = new MemoryStream();
encoder.Save(stream);
this.LayoutTransform = transform;
return new Bitmap(stream);
}