private void ProcessNetwork()
{
app.WriteLine("Downsampling images...");
foreach (ImagePair pair in imageList)
{
IMLData ideal = new BasicMLData(outputCount);
int idx = pair.Identity;
for (int i = 0; i < outputCount; i++)
{
if (i == idx)
{
ideal.Data[i] = 1;
}
else
{
ideal.Data[i] = -1;
}
}
try
{
var img = new Bitmap(pair.File);
var data = new ImageMLData(img);
training.Add(data, ideal);
}
catch (Exception e)
{
app.WriteLine("Error loading: " + pair.File
+ ": " + e.Message);
}
}
String strHidden1 = GetArg("hidden1");
String strHidden2 = GetArg("hidden2");
if (training.Count == 0)
{
app.WriteLine("No images to create network for.");
return;
}
training.Downsample(downsampleHeight, downsampleWidth);
int hidden1 = int.Parse(strHidden1);
int hidden2 = int.Parse(strHidden2);
network = EncogUtility.SimpleFeedForward(training
.InputSize, hidden1, hidden2,
training.IdealSize, true);
app.WriteLine("Created network: " + network);
}
private void Learn_Click(object sender, RoutedEventArgs e)
{
var downsample = new Downsampler();
var training = new ImageMLDataSet(downsample, true, 1, -1);
for (var i = 0; i < Images.Count; ++i)
{
var ideal = new BasicMLData(DIGITS_COUNT);
for (int j = 0; j < DIGITS_COUNT; ++j)
{
if (j == i)
{
ideal[j] = 1;
}
else
{
ideal[j] = -1;
}
}
foreach (var img in Images[i])
{
MemoryStream stream = new MemoryStream();
BitmapEncoder encoder = new BmpBitmapEncoder();
encoder.Frames.Add(BitmapFrame.Create(img));
encoder.Save(stream);
var bitmap = new Drawing.Bitmap(stream);
var data = new ImageMLData(bitmap);
training.Add(data, ideal);
}
}
training.Downsample(DIGIT_HEIGHT, DIGIT_WIDTH);
network = EncogUtility.SimpleFeedForward(training.InputSize, 35, 0, training.IdealSize, true);
double strategyError = 0.01;
int strategyCycles = 2000;
var train = new ResilientPropagation(network, training);
//train.AddStrategy(new ResetStrategy(strategyError, strategyCycles));
EncogUtility.TrainDialog(train, network, training);
EncogDirectoryPersistence.SaveObject(new FileInfo("network.eg"), network);
}
public void ProcessWhatIs()
{
String filename = GetArg("image");
try
{
var img = new Bitmap(filename);
var input = new ImageMLData(img);
input.Downsample(downsample, false, downsampleHeight,
downsampleWidth, 1, -1);
int winner = network.Winner(input);
app.WriteLine("What is: " + filename + ", it seems to be: "
+ neuron2identity[winner]);
}
catch (Exception e)
{
app.WriteLine("Error loading: " + filename + ", " + e.Message);
}
}
private void button1_Click(object sender, EventArgs e)
{
using (OpenFileDialog dlg = new OpenFileDialog())
{
dlg.Title = "Open Image";
dlg.Filter = "Image files (*.bmp, *.jpg, *.jpeg, *.jpe, *.jfif, *.png) | *.bmp; *.jpg; *.jpeg; *.jpe; *.jfif; *.png";
if (dlg.ShowDialog() == DialogResult.OK)
{
var bitMap = new Bitmap(dlg.FileName);
NeuralForm n = new NeuralForm();
n.BitMap = bitMap;
n.Show();
//
var network = new BasicNetwork();
FileInfo networkFile = new FileInfo(NETWORK_PATH);
if (System.IO.File.Exists(NETWORK_PATH))
{
network = (BasicNetwork)(Encog.Persist.EncogDirectoryPersistence.LoadObject(networkFile));
}
int hCounter = 0; int Wcounter = 0;
int hMax = bitMap.Height / SIZE;
int wMax = bitMap.Width / SIZE;
var outputBitmap = new Bitmap(bitMap.Width, bitMap.Height);
while (hMax > hCounter)
{
Encog.ML.Data.Image.ImageMLDataSet testingSet = new Encog.ML.Data.Image.ImageMLDataSet(new Encog.Util.DownSample.RGBDownsample(), false, 1, -1);
var _25x25bitmap = new Bitmap(SIZE, SIZE);
for (int i = 0; i < SIZE; i++)
{
for (int j = 0; j < SIZE; j++)
{
var px = bitMap.GetPixel(Wcounter * SIZE + j, hCounter * SIZE + i);
_25x25bitmap.SetPixel(i, j, px);
}
}
ImageMLData data = new ImageMLData(_25x25bitmap);
testingSet.Add(data, null);
testingSet.Downsample(SIZE, SIZE);
bool isVege = false;
foreach (IMLDataPair pair in testingSet)
{
IMLData output = network.Compute(pair.Input);
if (output[1] > .05)
{
isVege = true;
}
}
for (int i = 0; i < SIZE; i++)
{
for (int j = 0; j < SIZE; j++)
{
var px = _25x25bitmap.GetPixel(i, j);
if (!isVege)
{
outputBitmap.SetPixel(Wcounter * SIZE + j, hCounter * SIZE + i, px);
}
else
{
outputBitmap.SetPixel(Wcounter * SIZE + j, hCounter * SIZE + i, Color.FromArgb(px.R, 0, px.B));
}
}
}
Wcounter++;
if (wMax <= Wcounter)
{
Wcounter = 0;
hCounter++;
}
}
NeuralForm n2 = new NeuralForm();
n2.BitMap = outputBitmap;
n2.Show();
}
}
}
private void RecCanvas_MouseUp(object sender, MouseButtonEventArgs e)
{
recCanvas_drawing = false;
var size = new Size(RecCanvas.ActualWidth, RecCanvas.ActualHeight);
// Measure and arrange the surface
// VERY IMPORTANT
RecCanvas.Measure(size);
RecCanvas.Arrange(new Rect(size));
var renderBitmap = new RenderTargetBitmap((int)size.Width, (int)size.Height, 96d, 96d, PixelFormats.Pbgra32);
renderBitmap.Render(RecCanvas);
MemoryStream stream = new MemoryStream();
BitmapEncoder encoder = new BmpBitmapEncoder();
encoder.Frames.Add(BitmapFrame.Create(renderBitmap));
encoder.Save(stream);
var bitmap = new Drawing.Bitmap(stream);
var downsample = new Downsampler();
var result = downsample.DownSample(bitmap, DIGIT_HEIGHT, DIGIT_WIDTH);
var image = new Drawing.Bitmap(DIGIT_WIDTH, DIGIT_HEIGHT);
for (var i = 0; i < DIGIT_HEIGHT; ++i)
{
for (var j = 0; j < DIGIT_WIDTH; ++j)
{
image.SetPixel(j, i, Drawing.Color.FromArgb(
255,
(int)result[i * DIGIT_WIDTH + j],
(int)result[i * DIGIT_WIDTH + j],
(int)result[i * DIGIT_WIDTH + j]
));
}
}
RecComp.Source = Imaging.CreateBitmapSourceFromHBitmap(
image.GetHbitmap(),
IntPtr.Zero,
Int32Rect.Empty,
BitmapSizeOptions.FromEmptyOptions());
if (network == null)
{
return;
}
var input = new ImageMLData(bitmap);
input.Downsample(downsample, false, DIGIT_HEIGHT, DIGIT_WIDTH, 1, -1);
int winner = network.Winner(input);
RecDigit.Text = winner.ToString();
var data = network.Compute(input);
for (var i = 0; i < DIGITS_COUNT; ++i)
{
Results[i] = new ResultObject
{
Char = i.ToString(),
Result = (data[i] + 1) / 2
};
}
}
public static void Run()
{
FileInfo networkFile = new FileInfo(@"D:\Imagery\network\network.eg");
var network = new BasicNetwork();
network.AddLayer(new BasicLayer(null, true, SIZE * SIZE * 3));
network.AddLayer(new BasicLayer(new ActivationSigmoid(), true, NERUONCOUNT));
network.AddLayer(new BasicLayer(new ActivationSigmoid(), false, 2));
network.Structure.FinalizeStructure();
network.Reset();
if (System.IO.File.Exists(@"D:\Imagery\network\network.eg"))
{
network = (BasicNetwork)(Encog.Persist.EncogDirectoryPersistence.LoadObject(networkFile));
}
Encog.ML.Data.Image.ImageMLDataSet trainingSet = new Encog.ML.Data.Image.ImageMLDataSet(new RGBDownsample(), false, 1, -1);
Random rnd = new Random();
//take 1000,, take 5000 from nothing and scramble them
List <string> fileEntries = Directory.GetFiles(@"D:\Imagery\_Vege").OrderBy(x => rnd.Next()).Take(1000).ToList();
fileEntries.AddRange(Directory.GetFiles(@"D:\Imagery\_Nothing").OrderBy(x => rnd.Next()).Take(5000).ToArray());
fileEntries = fileEntries.OrderBy(x => rnd.Next()).Take(6000).ToList();
foreach (var file in fileEntries)
{
var bitmap = new System.Drawing.Bitmap(file);
ImageMLData data = new ImageMLData(bitmap);
if (file.Contains("_Nothing"))
{
BasicMLData ideal = new BasicMLData(Nothing);
trainingSet.Add(data, ideal);
}
else
{
BasicMLData ideal = new BasicMLData(Vegetation);
trainingSet.Add(data, ideal);
}
}
trainingSet.Downsample(SIZE, SIZE);
IMLTrain train = new Backpropagation(network, trainingSet, .001, 0.02)
{
};
int epoch = 1;
do
{
train.Iteration();
Console.WriteLine(@"Epoch #" + epoch + @" Error: " + train.Error);
epoch++;
} while (epoch < 50);
train.FinishTraining();
Encog.Persist.EncogDirectoryPersistence.SaveObject(networkFile, (BasicNetwork)network);
Encog.ML.Data.Image.ImageMLDataSet testingSet = new Encog.ML.Data.Image.ImageMLDataSet(new RGBDownsample(), false, 1, -1);
fileEntries = Directory.GetFiles(@"D:\Imagery\_VegeTest").ToList();
foreach (var file in fileEntries)
{
ImageMLData data = new ImageMLData(new System.Drawing.Bitmap(file));
BasicMLData ideal = new BasicMLData(Vegetation);
testingSet.Add(data, ideal);
}
fileEntries = Directory.GetFiles(@"D:\Imagery\_NothingTest").ToList();
foreach (var file in fileEntries)
{
ImageMLData data = new ImageMLData(new System.Drawing.Bitmap(file));
BasicMLData ideal = new BasicMLData(Nothing);
testingSet.Add(data, ideal);
}
testingSet.Downsample(SIZE, SIZE);
Console.WriteLine(@"Neural Network Results:");
foreach (IMLDataPair pair in testingSet)
{
IMLData output = network.Compute(pair.Input);
Console.WriteLine(@", actual (" + output[0] + @"," + output[1] + @"),ideal (" + pair.Ideal[0] + @"," + pair.Ideal[1] + ")");
}
EncogFramework.Instance.Shutdown();
}
