static void Main(string[] args)
{
//var readLine = Console.ReadLine();
//if (readLine == null) return;
//var args = readLine.Split(' ');
//var args = new[] { "house2.bmp", "house222.bmp", "canny", "1", "0,25", "0,1" };
//task1
//Инверсия значений пикселей изображения
if (args.Length == 3 && args[2] == "invert")
{
string inputFileName = args[0], outputFileName = args[1];
if (!File.Exists(inputFileName))
{
return;
}
var image = ImageIO.FileToGrayscaleFloatImage(inputFileName);
Inversion.InversionProcess(image);
ImageIO.ImageToFile(image, outputFileName);
}
//Отражение изображения по вертикали и по горизонтали
if (args.Length == 4 && args[2] == "mirror")
{
string inputFileName = args[0], outputFileName = args[1];
if (!File.Exists(inputFileName))
{
return;
}
var image = ImageIO.FileToGrayscaleFloatImage(inputFileName);
if (args[3] == "x")
{
Reflection.FlipHorizontal(image);
}
if (args[3] == "y")
{
Reflection.FlipVertical(image);
}
ImageIO.ImageToFile(image, outputFileName);
}
//Поворот изображений по и против часовой стрелки на 90, 180 и 270 градусов(на произвольный угол)
if (args.Length == 5 && args[2] == "rotate")
{
string inputFileName = args[0], outputFileName = args[1];
if (!File.Exists(inputFileName))
{
return;
}
var image = ImageIO.FileToGrayscaleFloatImage(inputFileName);
if (args[3] == "cw")
{
image = Rotate.ClockWiseRotate(image, angle: int.Parse(args[4]));
}
if (args[3] == "ccw")
{
image = Rotate.CounterClockWiseRotate(image, angle: int.Parse(args[4]));
}
ImageIO.ImageToFile(image, outputFileName);
}
//фильтр Превитта
if (args.Length == 4 && args[2] == "prewitt")
{
string inputFileName = args[0], outputFileName = args[1];
if (!File.Exists(inputFileName))
{
return;
}
var image = ImageIO.FileToGrayscaleFloatImage(inputFileName);
if (args[3] == "x")
{
image = MedianFiltering.Convolution(image, Filters.PrewittHorizontal(), 3, 2);
}
if (args[3] == "y")
{
image = MedianFiltering.Convolution(image, Filters.PrewittVertical(), 3, 2);
}
ImageIO.ImageToFile(image, outputFileName);
}
//фильтр Собеля
if (args.Length == 4 && args[2] == "sobel")
{
string inputFileName = args[0], outputFileName = args[1];
if (!File.Exists(inputFileName))
{
return;
}
var image = ImageIO.FileToGrayscaleFloatImage(inputFileName);
if (args[3] == "x")
{
image = MedianFiltering.Convolution(image, Filters.SobelHorizontal(), 3);
}
if (args[3] == "y")
{
image = MedianFiltering.Convolution(image, Filters.SobelVertical(), 3);
}
ImageIO.ImageToFile(image, outputFileName);
}
//фильтр Робертса
if (args.Length == 4 && args[2] == "roberts")
{
string inputFileName = args[0], outputFileName = args[1];
if (!File.Exists(inputFileName))
{
return;
}
var image = ImageIO.FileToGrayscaleFloatImage(inputFileName);
if (args[3] == "1")
{
image = MedianFiltering.Convolution(image, Filters.RobertsMainDiagonal(), 2, 2);
}
if (args[3] == "2")
{
image = MedianFiltering.Convolution(image, Filters.RobertsAdditionalDiagonal(), 2, 2);
}
ImageIO.ImageToFile(image, outputFileName);
}
//Медианная фильтрация с квадратным окном произвольного размера
if (args.Length == 4 && args[2] == "median")
{
string inputFileName = args[0], outputFileName = args[1];
if (!File.Exists(inputFileName))
{
return;
}
var image = ImageIO.FileToGrayscaleFloatImage(inputFileName);
image = MedianFiltering.MedianFilter(image, kernelSize: int.Parse(args[3]));
ImageIO.ImageToFile(image, outputFileName);
}
//Свёртка с фильтром Гаусса с произвольным выбором параметра — радиуса σ
if (args.Length == 4 && args[2] == "gauss")
{
string inputFileName = args[0], outputFileName = args[1];
if (!File.Exists(inputFileName))
{
return;
}
var image = ImageIO.FileToGrayscaleFloatImage(inputFileName);
var data = new double[image.rawdata.Length];
var template = new double[image.rawdata.Length];
for (var i = 0; i < image.rawdata.Length; i++)
{
data[i] = Convert.ToDouble(image.rawdata[i]);
}
ConvolutionGauss.GaussProcess(data, image.Width, image.Height, sigma: double.Parse(args[3]), windowSize: 9, temp: template, dest: data);
for (var i = 0; i < image.rawdata.Length; i++)
{
image.rawdata[i] = Convert.ToSingle(data[i]);
}
ImageIO.ImageToFile(image, outputFileName);
}
//Вычисление модуля градиента как корень из суммы квадратов свёрток с первой производной фильтра Гаусса по горизонтали и вертикали
if (args.Length == 4 && args[2] == "gradient")
{
string inputFileName = args[0], outputFileName = args[1];
if (!File.Exists(inputFileName))
{
return;
}
var image = ImageIO.FileToGrayscaleFloatImage(inputFileName);
ConvolutionGauss.GradientProcess(image.rawdata, image.Width, image.Height, double.Parse(args[3]), (int)(double.Parse(args[3]) * 6), image.rawdata);
ImageIO.ImageToFile(image, outputFileName);
}
//Фильтр Габора с произвольными параметрами
if (args.Length == 8 && args[2] == "gabor")
{
string inputFileName = args[0], outputFileName = args[1];
if (!File.Exists(inputFileName))
{
return;
}
var image = ImageIO.FileToGrayscaleFloatImage(inputFileName);
image = ConvolutionGauss.Gabor(image, Filters.Gabor((int)(6 * double.Parse(args[3])), double.Parse(args[3]), double.Parse(args[6]), double.Parse(args[5]), double.Parse(args[4]), double.Parse(args[6])), (int)(6 * double.Parse(args[3])), (int)(6 * double.Parse(args[3])), int.Parse(args[7]));
ImageIO.ImageToFile(image, outputFileName);
}
//Обнаружение сосудов на изображениях глазного дна с помощью фильтров Габора
if (args.Length == 4 && args[2] == "vessels")
{
string inputFileName = args[0], outputFileName = args[1];
if (!File.Exists(inputFileName))
{
return;
}
var image = ImageIO.FileToGrayscaleFloatImage(inputFileName);
image = ConvolutionGauss.Vessels(image, (int)(6 * Convert.ToDouble(args[3])), Convert.ToDouble(args[3]));
ImageIO.ImageToFile(image, outputFileName);
}
//task2
//Увеличение изображений в вещественное число раз с помощью билинейной интерполяции
if (args.Length == 4 && args[2] == "up_bilinear")
{
string inputFileName = args[0], outputFileName = args[1];
if (!File.Exists(inputFileName))
{
return;
}
var image = ImageIO.FileToGrayscaleFloatImage(inputFileName);
image = ImageResolution.Bilinear(image, Convert.ToDouble(args[3]));
ImageIO.ImageToFile(image, outputFileName);
}
//Увеличение изображений в вещественное число раз с помощью бикубической интерполяции
if (args.Length == 4 && args[2] == "up_bicubic")
{
string inputFileName = args[0], outputFileName = args[1];
if (!File.Exists(inputFileName))
{
return;
}
var image = ImageIO.FileToGrayscaleFloatImage(inputFileName);
var resultImage = ImageResolution.Bicubic(image, Convert.ToDouble(args[3]));
ImageIO.ImageToFile(resultImage, outputFileName);
}
//Понижение разрешения изображений в вещественное число раз
if (args.Length == 4 && args[2] == "downsample")
{
string inputFileName = args[0], outputFileName = args[1];
if (!File.Exists(inputFileName))
{
return;
}
var image = ImageIO.FileToGrayscaleFloatImage(inputFileName);
image = ImageResolution.DownBilinear(image, Convert.ToDouble(args[3]));
ImageIO.ImageToFile(image, outputFileName);
}
//Вычисление метрик сравнения изображений(MSE и PSNR, SSIM и MSSIM)
if (args.Length == 4 && args[2] == "metric")
{
string inputFileName = args[0], inputFileName2 = args[1];
if (!File.Exists(inputFileName) || !File.Exists(inputFileName2))
{
return;
}
var image = ImageIO.FileToGrayscaleFloatImage(inputFileName);
var image2 = ImageIO.FileToGrayscaleFloatImage(inputFileName2);
double result;
switch (args[3])
{
case "mse":
result = Metrics.Mse(image, image2);
Console.WriteLine(result);
Console.ReadKey();
break;
case "psnr":
result = Metrics.Psnr(image, image2);
Console.WriteLine(result);
Console.ReadKey();
break;
case "ssim":
result = Metrics.Ssim(image.rawdata, image2.rawdata, image.Width, image.Height);
Console.WriteLine(result);
Console.ReadKey();
break;
case "mssim":
result = Metrics.Mssim(image.rawdata, image2.rawdata, image.Width, image.Height);
Console.WriteLine(result);
Console.ReadKey();
break;
}
}
//task3
//Алгоритм детектирования границ Канни
if (args.Length == 6 && args[2] == "canny")
{
string inputFileName = args[0], outputFileName = args[1];
if (!File.Exists(inputFileName))
{
return;
}
var image = ImageIO.FileToGrayscaleFloatImage(inputFileName);
var image2 = Canny.Process(image, Convert.ToSingle(args[3]), Convert.ToSingle(args[4]), Convert.ToSingle(args[5]));
ImageIO.ImageToFile(image2, outputFileName);
}
//Алгоритм Харриса для детектирования углов
if (args.Length == 4 && args[2] == "harris")
{
string inputFileName = args[0], outputFileName = args[1];
if (!File.Exists(inputFileName))
{
return;
}
var image = ImageIO.FileToGrayscaleFloatImage(inputFileName);
image = Harris.Process(image, double.Parse(args[3]));
ImageIO.ImageToFile(image, outputFileName);
}
//Билатеральная фильтрация изображений
if (args.Length == 5 && args[2] == "bilateral")
{
string inputFileName = args[0], outputFileName = args[1];
if (!File.Exists(inputFileName))
{
return;
}
var image = ImageIO.FileToGrayscaleFloatImage(inputFileName);
image = Bilateral.Process(image, double.Parse(args[3]), double.Parse(args[3]));
ImageIO.ImageToFile(image, outputFileName);
}
}
public static Bitmap SmoothingFilter(this Bitmap sourceBitmap, MainForm form,
SmoothingFilterType smoothFilter =
SmoothingFilterType.Nenhum)
{
Bitmap inputBitmap = null;
//Progress bar
form.algorithmProgress = 0;
switch (smoothFilter)
{
case SmoothingFilterType.Nenhum:
{
inputBitmap = sourceBitmap;
} break;
case SmoothingFilterType.Gaussiano3x3:
{
inputBitmap = sourceBitmap.ConvolutionFilter(
Matrix.Gaussian3x3, 1.0 / 16.0, 0);
} break;
case SmoothingFilterType.Gaussiano5x5:
{
inputBitmap = sourceBitmap.ConvolutionFilter(
Matrix.Gaussian5x5, 1.0 / 159.0, 0);
} break;
case SmoothingFilterType.Gaussiano7x7:
{
inputBitmap = sourceBitmap.ConvolutionFilter(
Matrix.Gaussian7x7, 1.0 / 136.0, 0);
} break;
case SmoothingFilterType.Mediano3x3:
{
inputBitmap = sourceBitmap.MedianFilter(3);
} break;
case SmoothingFilterType.Mediano5x5:
{
inputBitmap = sourceBitmap.MedianFilter(5);
} break;
case SmoothingFilterType.Mediano7x7:
{
inputBitmap = sourceBitmap.MedianFilter(7);
} break;
case SmoothingFilterType.Mediano9x9:
{
inputBitmap = sourceBitmap.MedianFilter(9);
} break;
case SmoothingFilterType.Mean3x3:
{
inputBitmap = sourceBitmap.ConvolutionFilter(
Matrix.Mean3x3, 1.0 / 9.0, 0);
} break;
case SmoothingFilterType.Mean5x5:
{
inputBitmap = sourceBitmap.ConvolutionFilter(
Matrix.Mean5x5, 1.0 / 25.0, 0);
} break;
case SmoothingFilterType.LowPass3x3:
{
inputBitmap = sourceBitmap.ConvolutionFilter(
Matrix.LowPass3x3, 1.0 / 16.0, 0);
} break;
case SmoothingFilterType.LowPass5x5:
{
inputBitmap = sourceBitmap.ConvolutionFilter(
Matrix.LowPass5x5, 1.0 / 60.0, 0);
} break;
case SmoothingFilterType.Sharpen3x3:
{
inputBitmap = sourceBitmap.ConvolutionFilter(
Matrix.Sharpen3x3, 1.0 / 8.0, 0);
} break;
}
//Progress bar
form.algorithmProgress = 20;
// START additional filters ADDED BY GABRIEL
inputBitmap = AForge.Imaging.Image.Clone(inputBitmap, PixelFormat.Format24bppRgb); //Accepted format
Bilateral filterB = new Bilateral();
Grayscale filterG = new Grayscale(0.2125, 0.7154, 0.0721); //arbitrary values
CannyEdgeDetector filterE = new CannyEdgeDetector();
ColorImageQuantizer filterC = new ColorImageQuantizer(new MedianCutQuantizer());
Dilatation filterD = new Dilatation();
//Bilateral filter as present in the article
filterB.KernelSize = form.kernelValue;
filterB.SpatialFactor = form.spatialFactor;
filterB.ColorFactor = form.colorFactor;
filterB.ColorPower = form.colorPower;
filterB.ApplyInPlace(inputBitmap);
form.algorithmProgress = 40;
/* GENERATING BORDERS */
//Generate a grayscale version for edge detection
Bitmap edges = filterG.Apply(inputBitmap);
filterE.HighThreshold = form.highThreshold;
filterE.LowThreshold = form.lowThreshold;
filterE.ApplyInPlace(edges); // a new image, edges, is created here defining the edges of inputBitmap
//Dilatation filter
edges = filterD.Apply(edges);
generateBorder(edges);
//Making bg transparent
edges.MakeTransparent(Color.White);
form.algorithmProgress = 70;
// Color reduction as present in the article
inputBitmap = filterC.ReduceColors(inputBitmap, form.colorReductionFactor); // reduces to 24 variation
inputBitmap = AForge.Imaging.Image.Clone(inputBitmap, PixelFormat.Format32bppArgb); //Accepted format
// images merge
Bitmap bitmapResult = new Bitmap(inputBitmap.Width, inputBitmap.Height, inputBitmap.PixelFormat);
Graphics g = Graphics.FromImage(bitmapResult);
g.DrawImage(inputBitmap, 0, 0, inputBitmap.Width, inputBitmap.Height);
g.DrawImage(edges, 0, 0, inputBitmap.Width, inputBitmap.Height);
// END additional filters ADDED BY GABRIEL
form.algorithmProgress = 100;
return(bitmapResult); // it was returning input Bitmap before
}
