public void DistBetwTwoClasses(int n1, int n2, double[][] arr, List <Myimage> temp1, List <Myimage> temp2)
{
for (int j = 0; j < n1; j++)
{
for (int k = 0; k < n2; k++)
{
{ arr[j][k] = PerceptiveHash.hamming(temp1[j].Hash, temp2[k].Hash); }
}
}
}
public void DistAtOneClass(int n, double[][] arr, List <Myimage> temp)
{
for (int j = 0; j < n; j++)
{
for (int k = 0; k < n; k++)
{
{ if (j != k)
{
arr[j][k] = PerceptiveHash.hamming(temp[j].Hash, temp[k].Hash);
}
}
}
}
}
public string Recognize(Myimage X)
{
Myimage k = new Myimage();
double min = PerceptiveHash.hamming(X.Hash, etal1[0].Hash);
for (int i = 0; i < etal1.Count; i++)
{
Int64 h = PerceptiveHash.hamming(X.Hash, etal1[i].Hash);
if (min >= h)
{
min = h; k = etal1[i];
}
}
return(k.DirectoryPath);
}
public void CheckHash(RadioButton rb1, RadioButton rb2, RadioButton rb3, RadioButton rb4, BitmapImage bm1, Myimage m)
{
if (rb1.IsChecked == true)
{
m.Hash = PerceptiveHash.aHash(bm1);
}
if (rb2.IsChecked == true)
{
m.Hash = PerceptiveHash.pHash(bm1);
}
if (rb3.IsChecked == true)
{
m.Hash = PerceptiveHash.dHash(bm1);
}
if (rb4.IsChecked == true)
{
m.Hash = PerceptiveHash.gHash(bm1);
}
}
public void FindGroup()
{
rez = new List <Int64>();
string s = Interaction.InputBox("Enter % of similarity");
double perc = int.Parse(s);
perc = Math.Round(perc * 0.64);
ListOfClusters.Clear();
for (int k = 0; k < imglist.Count; k++)
{
ListOfClusters.Add(new Cluster());
}
List <Myimage> tmp = new List <Myimage>();
for (int i = 0; i < imglist.Count; i++)
{
prbar.Dispatcher.Invoke(() => prbar.Value++, DispatcherPriority.Background);
int p = 0;
for (int k = 0; k < ListOfClusters.Count; k++)
{
if (ListOfClusters[k].ImgInCluster.Contains(imglist[i]))
{
p++;
}
}
if (p == 0)
{
ListOfClusters[i].ImgInCluster.Add(imglist[i]); tmp.Add(imglist[i]);
}
for (int j = 0; j < imglist.Count; j++)
{
Int64 ham = PerceptiveHash.hamming(imglist[i].Hash, imglist[j].Hash);
if ((ham <= (64 - perc)) && (i != j) && (!tmp.Contains(imglist[j])))
{
ListOfClusters[i].ImgInCluster.Add(imglist[j]);
tmp.Add(imglist[j]);
}
}
}
}
public void FindSimilar(Myimage m)
{
wrp.Children.Clear();
string s = Interaction.InputBox("Enter % of similarity");
double perc = int.Parse(s);
perc = Math.Round(perc * 0.64);
List <Myimage> similarList = new List <Myimage>();
for (int j = 0; j < imglist.Count; j++)
{
Int64 ham = PerceptiveHash.hamming(m.Hash, imglist[j].Hash);
if (ham <= (64 - perc))
{
similarList.Add(imglist[j]);
}
}
for (int i = 0; i < similarList.Count; i++)
{
Uri uriSource = new Uri(similarList[i].ImagePath);
bm1 = new BitmapImage();
bm1.BeginInit();
bm1.UriSource = uriSource;
bm1.EndInit();
System.Windows.Controls.Image im = new System.Windows.Controls.Image();
im.Source = bm1; im.Width = 100; im.Height = 75;
im.MouseEnter += new MouseEventHandler(OnMouseEnterHandler);
im.MouseLeave += new MouseEventHandler(OnMouseLeaveHandler);
ToolTip tooltip = new ToolTip {
Content = similarList[i].ImagePath
};
im.ToolTip = tooltip;
im.Margin = new Thickness(5, 5, 5, 5);
wrp.Children.Add(im); scrollv.ScrollToBottom();
}
}
public void clusterDend()
{
ListOfClusters.Clear();
iterations = new List <Iteration>();
// ListOfObj = new List<Myimage>();
// ListOfObj = imglist.GetRange(0, imglist.Count);
for (int k = 0; k < imglist.Count; k++)
{
ListOfClusters.Add(new Cluster());
ListOfClusters[k].ImgInCluster.Add(imglist[k]);
}
n = 1;
double[,] data = new double[imglist.Count, imglist.Count];
for (int i = 0; i < imglist.Count; i++)
{
for (int j = 0; j < imglist.Count; j++)
{
data[i, j] = PerceptiveHash.hamming(imglist[i].Hash, imglist[j].Hash);
}
}
RowsCount = imglist.Count;
ColumnsCount = imglist.Count;
List <List <double> > matrix = new List <List <double> >(RowsCount + 1);
for (int i = 0; i < RowsCount; i++)
{
var list = new List <double>(ColumnsCount);
for (int j = 0; j < ColumnsCount; j++)
{
list.Add(data[i, j]);
}
matrix.Add(list);
}
var list1 = new List <double>(ColumnsCount);
for (int j = 0; j < ColumnsCount; j++)
{
list1.Add(j);
}
matrix.Add(list1);
order = new List <List <int> >();
for (int i = 0; i < imglist.Count; i++)
{
order.Add(new List <int>());
order[i].Add(i);
}
orderlist = new List <Order>();
max = 0;
iterations.Add(new Iteration {
Clusters = Copy2(ListOfClusters)
});
while (ListOfClusters.FindAll(i => (i.ImgInCluster.Count != 0)).Count != n)
{
min = MinValue(matrix, out index1, out index2);
if (min > max)
{
max = min;
}
double d1 = 0; double d2 = 0;
for (int i = 0; i < ColumnsCount; i++)
{
if (index1 != i)
{
d1 = matrix[index1][i];
d2 = matrix[index2][i];
matrix[index1][i] = (d1 + d2) / 2;
matrix[i][index1] = (d1 + d2) / 2;
}
}
orderlist.Add(new Hashes.Order {
set1 = Copy(order[Convert.ToInt32(matrix[RowsCount][index1])]), set2 = Copy(order[Convert.ToInt32(matrix[RowsCount][index2])]), dist = min
});
order[Convert.ToInt32(matrix[RowsCount][index1])].AddRange(order[Convert.ToInt32(matrix[RowsCount][index2])]);
order[Convert.ToInt32(matrix[RowsCount][index2])].Clear();
ListOfClusters[Convert.ToInt32(matrix[RowsCount][index1])].ImgInCluster.AddRange(ListOfClusters[Convert.ToInt32(matrix[RowsCount][index2])].ImgInCluster);
ListOfClusters[Convert.ToInt32(matrix[RowsCount][index2])].ImgInCluster.Clear();
iterations.Add(new Iteration {
Clusters = Copy2(ListOfClusters)
});
RemoveRow(index2, matrix);
RemoveColumn(index2, matrix);
}
PaintDend();
}