You have seen that image processing happens up to 30 frames per second. This means that memory allocation and clean-up is happening around 30 times per second. This makes performance trail, though it does not matter for a small application; but for a more complex application where there is major work involved other than only a color camera, it could be a big concern.

To make your application perform better, the alternative way is to process images using the WriteableBitmap object, which serves the purpose of frequently updating the image pixel. You can find this WriteableBitmap object in the System.Windows.Media.Imaging namespace. The WriteableBitmap object works in a different way than BitmapSource. The WriteableBitmap object allocates the memory at once and updates only the pixel data on frame change. WriteableBitmap improves the performance by reducing the memory consumption as well as memory allocation and deallocation.

The overall implementation approach for WriteableBitmap can be performed very easily. First create the WriteableBitmap object:

private WriteableBitmap writeableBitmap;

Then enable the ColorStream channel and initialize the writeableBitmap object with the frame width, height, and pixel formats. Then assign the writeableBitmap object to the source property of image control. Then attach the ColorFrameReady event:

if (this.sensor !=null & this.sensor.IsRunning && !this.sensor.ColorStream.IsEnabled)
{
  this.sensor.ColorStream.Enable();
  this.writeableBitmap = new 
  WriteableBitmap(this.sensor.ColorStream.FrameWidth,
  this.sensor.ColorStream.FrameHeight, 96, 96,
  PixelFormats.Bgr32, null);
  VideoControl.Source = this.writeableBitmap;
  this.sensor.ColorFrameReady += sensor_ColorFrameReady;
}

Use the following code block for the ColorFrameReady event handler. The only difference you will find in this event handler is the WritePixels() method, which is being updated with the pixel data for every frame.

void sensor_ColorFrameReady(object sender, ColorImageFrameReadyEventArgs e)
{
  using (ColorImageFrame imageFrame = e.OpenColorImageFrame())
  {
    if (imageFrame != null)
    {
      byte[] pixelData = new byte[imageFrame.PixelDataLength];
      imageFrame.CopyPixelDataTo(pixelData);
      int stride = imageFrame.Width * imageFrame.BytesPerPixel;
      this.writeableBitmap.WritePixels(
      new Int32Rect(0, 0, this.writeableBitmap.PixelWidth, 
      this.writeableBitmap.PixelHeight),
      pixelData,
      stride,
      0);
    }
  }
}

You can see that with the WritePixels() method, the reference to writeableBitmap is getting updated with the new frame from the sensor. This is just updating the frame's pixels and keeping other things unchanged.