public static double EvaluateW(UnmanagedImage image)
{
byte[,] pixels = image.GetPixels();
double aggregate = 0;
int width = pixels.GetLength(0);
int height = pixels.GetLength(1);
for (int i = 0; i < width; i++)
{
for (int j = 0; j < height; j++)
{
aggregate += pixels[i, j];
}
}
aggregate = Math.Pow(width * height, -1) * aggregate;
double aggregate2 = 0;
for (int i = 0; i < width; i++)
{
for (int j = 0; j < height; j++)
{
aggregate2 += Math.Pow((pixels[i, j] - aggregate), 2);
}
}
return((4D / (width * height * Math.Pow(255, 2))) * aggregate2);
}
private static Bitmap FusionProcessing(Bitmap image,
IEnumerable <AlgorithmInfo> algorithms,
IEnumerable <EvaluationResult> algoritmsScores)
{
double totalScore = algoritmsScores.Sum(x => x.AlgorithScore);
var results = new Dictionary <double, byte[, ]>();
foreach (var score in algoritmsScores)
{
AlgorithmInfo algorithmInfo = algorithms.Where(x => x.CustomName == score.AlgorithCustomName).Single();
var algorithm = AppFacade.DI.Container.Resolve <IAlgorithm>(algorithmInfo.AlgorithName);
algorithm.SetParameters(algorithmInfo.Parameters);
algorithm.Input = new AlgorithmInput(image);
UnmanagedImage result = algorithm.ProcessData().Image;
results.Add(score.AlgorithScore / totalScore, result.GetPixels());
}
UnmanagedImage fusionImage =
UnmanagedImage.FromManagedImage(new Bitmap(image.Width, image.Height, image.PixelFormat));
var fusionBytes = new byte[image.Width, image.Height];
for (int i = 0; i < image.Width; i++)
{
for (int j = 0; j < image.Height; j++)
{
double level = results.Sum(keyValuePair => (keyValuePair.Key * keyValuePair.Value[i, j]));
fusionBytes[i, j] = (byte)level;
}
}
fusionImage.SetPixels(fusionBytes);
return(fusionImage.ToManagedImage());
}
public static double EvaluateM(UnmanagedImage image)
{
byte[,] pixels = image.GetPixels();
var stats = new ImageStatistics(image);
var minGrayLevel = (byte)stats.Gray.Min;
var maxGrayLevel = (byte)stats.Gray.Max;
return((double)(maxGrayLevel - minGrayLevel) / (double)(maxGrayLevel + minGrayLevel));
}
public static double EvaluateLocallyW(UnmanagedImage image)
{
byte[,] pixels = image.GetPixels();
int width = pixels.GetLength(0);
int height = pixels.GetLength(1);
double aggregate = 0;
for (int i = 0; i < width - 1; i++)
{
for (int j = 0; j < height; j++)
{
aggregate += Math.Pow((pixels[i, j] - pixels[i + 1, j]), 2);
}
}
for (int i = 0; i < width; i++)
{
for (int j = 0; j < height - 1; j++)
{
aggregate += Math.Pow((pixels[i, j] - pixels[i, j + 1]), 2);
}
}
for (int i = 0; i < width - 1; i++)
{
for (int j = 0; j < height - 1; j++)
{
aggregate += Math.Pow((pixels[i, j] - pixels[i + 1, j + 1]), 2);
}
}
for (int i = 0; i < width - 1; i++)
{
for (int j = 1; j < height; j++)
{
aggregate += Math.Pow((pixels[i, j] - pixels[i + 1, j - 1]), 2);
}
}
return((1D / (width * height * Math.Pow(255, 2))) * aggregate);
}
public static double EvaluateS(UnmanagedImage image)
{
byte[,] pixels = image.GetPixels();
var stats = new ImageStatistics(image);
var minGrayLevel = (byte)stats.Gray.Min;
var maxGrayLevel = (byte)stats.Gray.Max;
double aggregate = 0;
int width = pixels.GetLength(0);
int height = pixels.GetLength(1);
for (int i = 0; i < width; i++)
{
for (int j = 0; j < height; j++)
{
aggregate += pixels[i, j];
}
}
aggregate = Math.Pow(width * height, -1) * aggregate;
return((maxGrayLevel - minGrayLevel) / aggregate);
}