Question

In Java i would do something like this

int[] pixels = ((DataBufferInt) image.getRaster().getDataBuffer()).getData();

where image is a BufferedImage and then alter the pixels in there and make my own blitting methods but how am i supposed to do something like this in C#? I know i can use a Bitmap to replace BufferedImage in C# but i’m not sure about making a reference to the data like shown above.

Answer 1

You would use Lockbits and Marshal.Copy:

Rectangle rect = new Rectangle(0, 0, bmp.Width, bmp.Height);
System.Drawing.Imaging.BitmapData bmpData =
  bmp.LockBits(rect, System.Drawing.Imaging.ImageLockMode.ReadWrite, bmp.PixelFormat);

IntPtr ptr = bmpData.Scan0;
int bytes  = Math.Abs(bmpData.Stride) * bmp.Height;
byte[] rgbValues = new byte[bytes];

// Copy the RGB values into the array.
System.Runtime.InteropServices.Marshal.Copy(ptr, rgbValues, 0, bytes);

// do something with the array

// Copy the RGB values back to the bitmap
System.Runtime.InteropServices.Marshal.Copy(rgbValues, 0, ptr, bytes);

bmp.UnlockBits(bmpData);

Note: The code is basically the example code from the LockBits documentation page, but the code has a limitation. It assumes that the Stride value is positive, i.e. that the image is not stored upside down in memory, eventhough the use of Math.Abs on the Stride value indicates that the person writing the code was aware that the Stride value can be negative.

For a negative Stride value the Scan0 can not be used as the starting address of a continuos memory block, as it’s the address of the first scan line. The starting address of the memory block would be the starting address of the last line in the image, not the first.

That address would be bmpData.Scan0 + bmpData.Stride * (bmp.Height - 1)