4 files changed, 237 insertions, 0 deletions

diff --git a/SD-VBS/common/toolbox/toolbox_basic/textons/dist2.m b/SD-VBS/common/toolbox/toolbox_basic/textons/dist2.m
new file mode 100755
index 0000000..f2d93e1
--- /dev/null
+++ b/SD-VBS/common/toolbox/toolbox_basic/textons/dist2.m
@@ -0,0 +1,27 @@
+function n2 = dist2(x, c)
+%DIST2  Calculates squared distance between two sets of points.
+%
+%       Description
+%       D = DIST2(X, C) takes two matrices of vectors and calculates the
+%       squared Euclidean distance between them.  Both matrices must be of
+%       the same column dimension.  If X has M rows and N columns, and C has
+%       L rows and N columns, then the result has M rows and L columns.  The
+%       I, Jth entry is the  squared distance from the Ith row of X to the
+%       Jth row of C.
+%
+%       See also
+%       GMMACTIV, KMEANS, RBFFWD
+%
+
+%       Copyright (c) Christopher M Bishop, Ian T Nabney (1996, 1997)
+
+[ndata, dimx] = size(x);
+[ncentres, dimc] = size(c);
+if dimx ~= dimc
+        error('Data dimension does not match dimension of centres')
+end
+
+n2 = (ones(ncentres, 1) * sum((x.^2)', 1))' + ...
+                ones(ndata, 1) * sum((c.^2)',1) - ...
+                2.*(x*(c'));
+
diff --git a/SD-VBS/common/toolbox/toolbox_basic/textons/find_textons.m b/SD-VBS/common/toolbox/toolbox_basic/textons/find_textons.m
new file mode 100755
index 0000000..e7fa4b0
--- /dev/null
+++ b/SD-VBS/common/toolbox/toolbox_basic/textons/find_textons.m
@@ -0,0 +1,46 @@
+function [centers,label,post,d2]=find_textons(FIw,ncenters,centers_in,n_iter);
+% [centers,label,post,d2]=find_textons(FIw,ncenters,centers_in,n_iter);
+% 
+% find textons using kmeans for windowed portion FIw of filtered image 
+%
+% to start with centers pulled randomly from image, set centers_in=[]
+
+% define number of textons
+%ncenters=25;
+
+[N1,N2,N3]=size(FIw);
+% reshape filtered image stack into a long array of feature vectors
+fv=reshape(FIw,N1*N2,N3);
+% (each row is a feature vector)
+
+%centers=.01^2*randn(ncenters,N3);
+% take centers randomly from within image
+if isempty(centers_in)
+   rndnum=1+floor(N1*N2*rand(1,ncenters));
+   centers_in=fv(rndnum,:);
+end
+
+options = foptions;
+options(1)=1;           % Prints out error values.
+options(5) = 0;
+if nargin<4
+   n_iter=15;
+end
+options(14) = n_iter;           % Number of iterations.
+
+[centers,options,d2,post]=kmeans2(centers_in,fv,options);
+
+% reshuffle the centers so that the one closest to the origin
+% (featureless) comes last
+norms=sum(centers.^2,2);
+[sortval,sortind]=sort(-norms);
+centers=centers(sortind,:);
+d2=reshape(d2,N1,N2,ncenters);
+post=reshape(post,N1,N2,ncenters);
+d2=d2(:,:,sortind);
+post=post(:,:,sortind);
+
+
+% retrieve cluster number assigned to each feature vector
+[minval,label]=min(d2,[],3);
+
diff --git a/SD-VBS/common/toolbox/toolbox_basic/textons/find_textons1.m b/SD-VBS/common/toolbox/toolbox_basic/textons/find_textons1.m
new file mode 100755
index 0000000..b192015
--- /dev/null
+++ b/SD-VBS/common/toolbox/toolbox_basic/textons/find_textons1.m
@@ -0,0 +1,37 @@
+function [centers,label,post,d2]=find_textons(fv,ncenters,centers_in,n_iter);
+% [centers,label,post,d2]=find_textons(FIw,ncenters,centers_in,n_iter);
+% 
+% find textons using kmeans for windowed portion FIw of filtered image 
+%
+% to start with centers pulled randomly from image, set centers_in=[]
+
+[N1,N2] =size(fv);
+
+% take centers randomly from within image
+if isempty(centers_in)
+   rndnum=1+floor(N1*rand(1,ncenters));
+   centers_in=fv(rndnum,:);
+end
+
+options = foptions;
+options(1)=1;           % Prints out error values.
+options(5) = 0;
+if nargin<4
+   n_iter=15;
+end
+options(14) = n_iter;           % Number of iterations.
+
+[centers,options,d2,post]=kmeans2(centers_in,fv,options);
+
+
+% retrieve cluster number assigned to each feature vector
+[minval,label]=min(d2,[],2);
+
+
+h = hist(label(:),[1:max(label(:))]);
+a = h>0;
+a = cumsum(a);
+
+[nr,nc] = size(label);
+label = reshape(a(label(:)),nr,nc);
+
diff --git a/SD-VBS/common/toolbox/toolbox_basic/textons/kmeans2.m b/SD-VBS/common/toolbox/toolbox_basic/textons/kmeans2.m
new file mode 100755
index 0000000..0bd87b2
--- /dev/null
+++ b/SD-VBS/common/toolbox/toolbox_basic/textons/kmeans2.m
@@ -0,0 +1,127 @@
+function [centres, options, d2, post, errlog] = kmeans2(centres, data, options)
+%KMEANS Trains a k means cluster model.
+%
+%       Description
+%        CENTRES = KMEANS(CENTRES, DATA, OPTIONS) uses the batch K-means
+%       algorithm to set the centres of a cluster model. The matrix DATA
+%       represents the data which is being clustered, with each row
+%       corresponding to a vector. The sum of squares error function is used.
+%       The point at which a local minimum is achieved is returned as
+%       CENTRES.  The error value at that point is returned in OPTIONS(8).
+%
+%       [CENTRES, OPTIONS, POST, ERRLOG] = KMEANS(CENTRES, DATA, OPTIONS)
+%       also returns the cluster number (in a one-of-N encoding) for each
+%       data point in POST and a log of the error values after each cycle in
+%       ERRLOG.    The optional parameters have the following
+%       interpretations.
+%
+%       OPTIONS(1) is set to 1 to display error values; also logs error
+%       values in the return argument ERRLOG. If OPTIONS(1) is set to 0, then
+%       only warning messages are displayed.  If OPTIONS(1) is -1, then
+%       nothing is displayed.
+%
+%       OPTIONS(2) is a measure of the absolute precision required for the
+%       value of CENTRES at the solution.  If the absolute difference between
+%       the values of CENTRES between two successive steps is less than
+%       OPTIONS(2), then this condition is satisfied.
+%
+%       OPTIONS(3) is a measure of the precision required of the error
+%       function at the solution.  If the absolute difference between the
+%       error functions between two successive steps is less than OPTIONS(3),
+%       then this condition is satisfied. Both this and the previous
+%       condition must be satisfied for termination.
+%
+%       OPTIONS(14) is the maximum number of iterations; default 100.
+%
+%       See also
+%       GMMINIT, GMMEM
+%
+
+%       Copyright (c) Christopher M Bishop, Ian T Nabney (1996, 1997)
+
+[ndata, data_dim] = size(data);
+[ncentres, dim] = size(centres);
+
+if dim ~= data_dim
+  error('Data dimension does not match dimension of centres')
+end
+
+if (ncentres > ndata)
+  error('More centres than data')
+end
+
+% Sort out the options
+if (options(14))
+  niters = options(14);
+else
+  niters = 100;
+end
+
+store = 0;
+if (nargout > 3)
+  store = 1;
+  errlog = zeros(1, niters);
+end
+
+% Check if centres and posteriors need to be initialised from data
+if (options(5) == 1)
+  % Do the initialisation
+  perm = randperm(ndata);
+  perm = perm(1:ncentres);
+
+  % Assign first ncentres (permuted) data points as centres
+  centres = data(perm, :);
+end
+% Matrix to make unit vectors easy to construct
+id = eye(ncentres);
+
+% Main loop of algorithm
+for n = 1:niters
+
+  % Save old centres to check for termination
+  old_centres = centres;
+  
+  % Calculate posteriors based on existing centres
+  d2 = dist2(data, centres);
+  % Assign each point to nearest centre
+  [minvals, index] = min(d2', [], 1);
+  post = id(index,:);
+
+  num_points = sum(post, 1);
+  % Adjust the centres based on new posteriors
+  for j = 1:ncentres
+    if (num_points(j) > 0)
+      centres(j,:) = sum(data(find(post(:,j)),:), 1)/num_points(j);
+    end
+  end
+
+  % Error value is total squared distance from cluster centres
+  e = sum(minvals);
+  tmp = sort(minvals);
+  e95 = sqrt(tmp(round(length(tmp) * 0.95)));
+  erms = sqrt(e/ndata);
+  if store
+    errlog(n) = e;
+  end
+  if options(1) > 0
+    fprintf(1, '   Cycle %4d  RMS Error %11.6f  95-tier Error %11.6f\n', n, erms,e95);
+  end
+
+  if n > 1
+    % Test for termination
+    if max(max(abs(centres - old_centres))) < options(2) & ...
+        abs(old_e - e) < options(3)
+      options(8) = e;
+      return;
+    end
+  end
+  old_e = e;
+end
+
+% If we get here, then we haven't terminated in the given number of 
+% iterations.
+options(8) = e;
+%if (options(1) >= 0)
+%  disp('Warning: Maximum number of iterations has been exceeded');
+%end
+