/// <summary>
/// Unwarps omnidirectional image
/// </summary>
/// <param name="omniImage">The omnidirectional image</param>
/// <param name="centerX"></param>
/// <param name="centerY"></param>
/// <param name="blindSpotRadius"></param>
/// <param name="outerRadius"></param>
/// <returns></returns>
public static Color[,] Unwarp(Color[,] omniImage, int centerX, int centerY, int blindSpotRadius,
int outerRadius)
{
// Initialize a new perspective Image
int circumference = (int)Math.Round(2 * Math.PI * outerRadius);
Color[,] perspectiveImage = new Color[circumference, outerRadius - blindSpotRadius];
// Loop through coordinates of perspective image
for (int x = 0; x < perspectiveImage.GetLength(0) - 1; x++)
{
for (int y = 0; y < perspectiveImage.GetLength(1) - 1; y++)
{
// Convert to polar in omni
float normalizedX = (float)x / perspectiveImage.GetLength(0);
double angle = -2 * Math.PI * normalizedX;
float r = outerRadius - y;
// Convert polar to cartesian
double perspectiveX = r * Math.Cos(angle) + centerX;
double perspectiveY = r * Math.Sin(angle) + centerY;
// Interpolate the fractional coordinates
perspectiveImage[x, y] =
ImageScaler.Bilinear(omniImage, (float)Math.Abs(perspectiveX), (float)Math.Abs(perspectiveY));
}
}
return(perspectiveImage);
}
private static void Main(string[] args)
{
Color[,] forestImage = ImageViewer.LoadImage(ImagesForestJpg);
Color[,] omniImage1 = ImageViewer.LoadImage(Omni1Jpg);
Color[,] omniImage2 = ImageViewer.LoadImage(Omni2Jpg);
Color[,] omniImage3 = ImageViewer.LoadImage(Omni3Jpg);
Color[,] tulips = ImageViewer.LoadImage(TulipsPng);
Color[,] baboon = ImageViewer.LoadImage(BaboonPng);
Color[,] aurora = ImageViewer.LoadImage(AuroraJpg);
Color[,] flowers = ImageViewer.LoadImage(FlowersPng);
#region Assignment 1
ColorConverter colorConverter = new ColorConverter();
// Convert image to yuv and back
Console.WriteLine("Showing conversion to YUY and back");
YuvColor[,] yuvColors = colorConverter.ConvertRgbImageToYuv(forestImage);
Color[,] imageConvertedBack = colorConverter.ConvertYuvToRgbImage(yuvColors);
ImageViewer.DrawImagePair(forestImage, imageConvertedBack);
//Convert rgb to greyscale
Console.WriteLine("Showing RGB to greyscale");
Color[,] greyscaleImage = colorConverter.GreyscaleRgbImage(forestImage);
ImageViewer.DrawImagePair(forestImage, greyscaleImage);
// Convert yuv to greyscale and to rgb for display
Console.WriteLine("Showing RGB to YUV to greyscale");
YuvColor[,] greyscaleYuv = colorConverter.GreyscaleYuvColors(yuvColors);
Color[,] greyscaleImageYuv = colorConverter.ConvertYuvToRgbImage(greyscaleYuv);
ImageViewer.DrawImagePair(forestImage, greyscaleImageYuv);
#endregion
#region Assignment 2
// Resize images using nearest neighbour interpolation
Color[,] scaledDownImageNn = ImageScaler.Scale(ImageScaler.NearestNeighbour, baboon, .2f);
Color[,] scaledUpImageNn = ImageScaler.Scale(ImageScaler.NearestNeighbour, scaledDownImageNn, 5.0f);
Console.WriteLine("Showing downscaling using nearest neighbour interpolation");
ImageViewer.DrawImagePair(baboon, scaledDownImageNn);
Console.WriteLine("Showing upscaling using nearest neighbour interpolation");
ImageViewer.DrawImagePair(scaledDownImageNn, scaledUpImageNn);
// Resize images using bilinear interpolation
Color[,] scaledDownImageB = ImageScaler.Scale(ImageScaler.Bilinear, baboon, .2f);
Color[,] scaledUpImageB = ImageScaler.Scale(ImageScaler.Bilinear, scaledDownImageB, 5.0f);
Console.WriteLine("Showing downscaling using bilinear interpolation");
ImageViewer.DrawImagePair(baboon, scaledDownImageB);
Console.WriteLine("Showing upscaling using nearest bilinear interpolation");
ImageViewer.DrawImagePair(scaledDownImageB, scaledUpImageB);
#endregion */
#region Assignment 3
Console.WriteLine("Showing image segmentation using k-means clustering with 2 clusters");
Color[,] segmentedImage = ImageSegmenter.Segment(flowers, 2);
ImageViewer.DrawImagePair(flowers, segmentedImage);
Console.WriteLine("Showing image segmentation using k-means clustering with 8 clusters");
Color[,] segmentedImage2 = ImageSegmenter.Segment(tulips, 8);
ImageViewer.DrawImagePair(tulips, segmentedImage2);
Console.WriteLine("Showing image segmentation using k-means clustering with 6 clusters");
Color[,] segmentedImage3 = ImageSegmenter.Segment(baboon, 4);
ImageViewer.DrawImagePair(baboon, segmentedImage3);
#endregion
#region Assignment 4
// Unwarp and display
Console.WriteLine("Showing omnidirectional image unwarping of image 1");
Color[,] perspective1 = OmniUnwarper.Unwarp(omniImage1, 330, 240, 16, 236);
ImageViewer.DrawImagePair(omniImage1, perspective1);
Console.WriteLine("Showing omnidirectional image unwarping of image 2");
Color[,] perspective2 = OmniUnwarper.Unwarp(omniImage2, 240, 231, 81, 232);
ImageViewer.DrawImagePair(omniImage2, perspective2);
Console.WriteLine("Showing omnidirectional image unwarping of image 3");
Color[,] perspective3 = OmniUnwarper.Unwarp(omniImage3, 312, 239, 35, 227);
ImageViewer.DrawImagePair(omniImage3, perspective3);
#endregion
}
