Algorithms file level help

From BanghamLab
Jump to navigation Jump to search

C:\Users\AB\DArT_Toolshed\Algorithms\Bezier

bezlib_bezcurve.m

 function [varargout] = bezlib_bezcurve(varargin)

 Calculates the points along a Bezier curve.

 Inputs: 
  ctrl_pts - the control points for the Bezier
    t      - the parameterization (time in this case)

 Outputs: 
  bpts - the xy points along the Bezier curve defined by ctrl_pts at time t.


 Example: 
 x = linspace(-1,1,100);
 y = x.^3 + randn(size(x))/5;
 pts = [x(:), y(:)];
 cpts = bezlib_bezierfitlms('pts', pts, 'order', 4);
 bpts = bezlib_bezcurve('ctrl_pts', cpts, 't', [0:0.001:1]);

 See also: bezlib_bezierfitlms

 Dr. A. I. Hanna (2005)

 revised. J. Strasser 22/08/2007


bezlib_bezdisp.m

 function varargout = bezlib_bezdisp(varargin)

 Plots the result of bezlib_bezierfitlms

 Inputs:
  'pts' - the original noisy data points
  'ctrl_pts' - the control points generated by bezlib_bezierfitlms
  'parent' - the axis to draw to (default = gca)

 Outputs:
  'ph' - the plot handles to the data

 cpts = bezlib_bezierfitlms('verbose', 1);
 bezlib_plotbezslopes('cpts', cpts, 'parent', gca);

 See also: bezlib_bezierfitlms, bezlib_bezcurve

 Dr. A. I. Hanna (2007)

 revised. J. Strasser 22/08/2007


bezlib_bezierfitlms.m

 function [varargout] = bezlib_bezierfitlms(varargin)


 Calculates the control points for a Bezier curve given some data using a
 least mean square approach. Note it is important that the point be
 ordered in terms of a parameterization, i.e. arc length or time.
 Unordered points will result in a poor fit.

 Inputs:
  pts - the Nx2 data that the Bezier should fit to (default = 100 points for a cubic polynomial)
  order - the order of the Bezier curve (default = 3)
  verbose - display any output (default = 0)

 Outputs:
  cpts - a set of Mx2 control points


 Example:
 cla;
 bezlib_bezierfitlms('verbose', 1);

 Example:
 cla;
 x = linspace(-1,1,100);
 y = x.^3 + randn(size(x))/5;
 pts = [x(:), y(:)];
 cpts = bezlib_bezierfitlms('pts', pts, 'order', 4, 'verbose', 1);
 bezlib_plotbezslopes('cpts', cpts, 'parent', gca);

 Example:
 cla;
 x = linspace(-2,2,100);
 y = x.^2 + randn(size(x))/5;
 z = x + y;
 pts = [x(:), y(:), z(:)];
 cpts = bezlib_bezierfitlms('pts', pts, 'order', 4, 'verbose', 1);
 bezlib_plotbezslopes('cpts', cpts, 'parent', gca);

 See also: bezlib_bezcurve, bezlib_plotbezslopes, bezlib_bezdisp

 Dr. A. I. Hanna (2006)

 revised. J. Strasser 22/08/2007


bezlib_plotbezslopes.m

 function varargout = bezlib_plotbezslopes(varargin)

 Plots the slopes of the control points for a bezier curve

 Inputs:
  'cpts' - the control points
  'parent' - the axis to draw to (default = gca)

 Outputs:
  'ph' - the plot handles to the slopes

 cpts = bezlib_bezierfitlms('verbose', 1);
 bezlib_plotbezslopes('cpts', cpts, 'parent', gca);

 See also: bezlib_bezierfitlms, bezlib_bezcurve

 Dr. A. I. Hanna (2007)

 revised. J. Strasser 22/08/2007


C:\Users\AB\DArT_Toolshed\Algorithms\MeshGeneration

C:\Users\AB\DArT_Toolshed\Algorithms\MeshGeneration\Mesh2d v2.1

connectivity.m

  CONNECTIVITY: Assemble connectivity data for a triangular mesh.

 The edge based connectivity is built for a triangular mesh and the
 boundary nodes identified. This data should be useful when implementing
 FE/FV methods using triangular meshes.

  [e,te,et2,bnd] = connectivity(p,t);

  p   : Nx2 array of nodes coordinates, [x1,y1; x2,y2; etc]
  t   : Mx3 array of triangles as indices, [n11,n12,n13; n21,n22,n23; etc]

  e   : Kx2 array of unique mesh edges - [n11,n12; n21,n22; etc]
  te  : Mx3 array of triangles as indices into E, [e11,e12,e13; 
                                                   e21,e22,e23; etc]
  e2t : Kx2 array of triangle neighbours for unique mesh edges -
        [t11,t12; t21,t22; etc]. Each row has two entries corresponding to
        the triangle numbers associated with each edge in E. Boundary
        edges have e2t(i,2)=0.
  bnd : Nx1 logical array identifying boundary nodes. P(i,:) is a boundary
        node if BND(i)=TRUE.

 See also MESH2D, REFINE


copyright.m

 Mesh2d is a MATLAB toolbox designed to automatically generate quality 2D 
 unstructured triangular meshes based on linear geometry inputs.
 Mesh2d is Copyright (C) 2007 Darren Engwirda. 

 This program is free software; you can redistribute it and/or 
 modify it under the terms of the GNU General Public License as 
 published by the Free Software Foundation; either version 2 of 
 the License, or (at your option) any later version. 

 This program is distributed in the hope that it will be useful, 
 but WITHOUT ANY WARRANTY; without even the implied warranty of 
 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU 
 General Public License for more details. 
 
 You should have received a copy of the GNU General Public License 
 along with this program; if not, write to the Free Software 
 Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, 
 USA. 

 If you use Mesh2d in any program or publication, please inform and
 acknowledge its author Darren Engwirda (d_engwirda@hotmail.com). 


inpoly.m

  INPOLY: Point-in-polygon testing.

 Determine whether a series of points lie within the bounds of a polygon
 in the 2D plane. General non-convex, multiply-connected polygonal
 regions can be handled.

  SHORT SYNTAX:

   in = inpoly(p,node);

   p   : The points to be tested as an Nx2 array [x1 y1; x2 y2; etc].
   node: The vertices of the polygon as an Mx2 array [X1 Y1; X2 Y2; etc].
         The standard syntax assumes that the vertices are specified in
         consecutive order.

   in  : An Nx1 logical array with IN(i) = TRUE if P(i,:) lies within the
         region.

  LONG SYNTAX:

  [in,on] = inpoly(p,node,edge);

  edge: An Mx2 array of polygon edges, specified as connections between
        the vertices in NODE: [n1 n2; n3 n4; etc]. The vertices in NODE
        do not need to be specified in connsecutive order when using the
        extended syntax.

  on  : An Nx1 logical array with ON(i) = TRUE if P(i,:) lies on a
        polygon edge. (A tolerance is used to deal with numerical
        precision, so that points within a distance of
        eps^0.8*norm(node(:),inf) from a polygon edge are considered "on"
        the edge.

 EXAMPLE:

   polydemo;       % Will run a few examples

 See also INPOLYGON


mesh2d.m

  MESH2D: 2D unstructured triangular mesh generation.

 A 2D unstructured triangular mesh is generated based on a piecewise-
 linear geometry input. An iterative method is implemented to optimise 
 mesh quality. General multiply connected domains and element size 
 functions can be specified.

 Returns the final coordinates of the nodes P, and their triangulation T
 (with a counter-clockwise node ordering).

  SHORT SYNTAX:

  [p,t] = mesh2d(node);

 NODE defines the geometry nodes as an Nx2 array:

  node  = [x1 y1; x2 y2; etc], geometry nodes specified in consectutive
                               order, such that NODE(2,:) is joined with
                               NODE(1,:) etc.

 An element size function is automatically generated based on the 
 complexity of the geometry. Generally this produces meshes with the 
 fewest number of triangles.

  LONG SYNTAX:

  [p,t] = mesh2d(node,edge,hdata,options);

 Blank arguments can be passed using the empty placeholder "[]".

 EDGE defines the connectivity between the points in NODE as a list of
 edges:

   edge = [n1 n2; n2 n3; etc], connectivity between nodes to form
                               geometry edges. If EDGE is specified it is
                               not required that NODE be consectutive.

 HDATA is a structure containing user defined element size information. 
 HDATA can include the following fields:

  hdata.hmax  = h0;                   Max allowable global element size.
  hdata.edgeh = [e1,h1; e2,h2; etc];  Element size on specified geometry 
                                      edges.
  hdata.fun   = 'fun' or @fun;        User defined size function.
  hdata.args  = {arg1, arg2, etc};    Additional arguments for HDATA.FUN.

 Calls to user specified functions must accept vectorised input of the 
 form H = FUN(X,Y,ARGS{:}), where X,Y are the xy coordinates where the
 element size will be evaluated and ARGS are optional additional arguments 
 as passed by HDATA.ARGS.

 An automatic size function is always generated to ensure that the
 geometry is adequately resolved. The overall size function is the minimum
 of the user specified and automatic functions.

 OPTIONS is a structure array that allows some of the "tuning" parameters
 used in the solver to be modified:

   options.mlim   : The convergence tolerance. The maximum percentage 
                    change in edge length per iteration must be less than 
                    MLIM { 0.05, 5% }. 
   options.maxit  : The maximum allowable number of iterations { 20 }.
   options.dhmax  : The maximum allowable (relative) gradient in the size 
                    function { 0.3 }.
   options.output : Displays the mesh and the mesh statistics upon
                    completion { TRUE }.

 EXAMPLE:

   meshdemo                  % Will run the standard demos
   mesh_collection(n)        % Will run some additional demos

 See also REFINE, SMOOTHMESH, DELAUNAYN


mesh_collection.m

  MESH_COLLECTION: Collection of meshing examples from MESH2D users. 

  mesh_collection(n) will run the nth example.

  1. Simple square domain. Used for "driven cavity" CFD studies.

  2. Rectangular domain with circular hole. Used in thermally coupled CFD
     studies to examine the flow around a heated pipe.

  3. Rectangular domain with circular hole and user defined size
     functions. Used in a CFD study to examine vortex shedding about
     cylinders.

  4. Rectangular domain with 2 circular holes and user defined size
     functions. Used in a CFD study to examine the unsteady flow between
     cylinders.

  5. Rectangular domain with square hole and user defined size functions.
     Used in a CFD study to examine vortex shedding about square prisms.

  6. 3 element airfoil with user defined size functions and boundary layer
     size functions. Used in a CFD study to examin the lift/drag
     characteristics.

  7. U shaped domain.

  8. Rectangular domain with step. Used for "backward facing step" CFD
     studies.

  9. NACA airfoil with boundary layer size functions. Used in a CFD study
     to examine the lift/drag vs. alpha characteristics.

  10. Wavy channel from Kong Zour. Used in a CFD study to examine unsteady
      behaviour.

  11. Tray of glass beads from Falk Hebe. Used in a CFD study to examine the flow
      through past a collection of beads.

  12. Coastline data from Francisco Garcia. PLEASE NOTE! This is a very
      complex example and takes a bit of processing time (40 sec on my 
      machine).

 I am always looking for new meshes to add to the collection, if you would
 like to contribute please send me an email with an m-file description of
 the NODE, EDGE, HDATA and OPTIONS used to setup the mesh.

 Darren Engwirda    : 2006-2007
 Email              : d_engwirda@hotmail.com


meshdemo.m

 Demo function for mesh2d.

 Feel free to "borrow" any of the geometries for your own use.

 Example:

   meshdemo;       % Runs the demos

 Darren Engwirda - 2006


C:\Users\AB\DArT_Toolshed\Algorithms\MeshGeneration\Mesh2d v2.1\private

MyDelaunayn.m


checkgeometry.m


dist2poly.m


fixmesh.m

FIXMESH  Remove duplicated/unused nodes and fix element orientation.
   [P,T]=FIXMESH(P,T)


mytsearch.m


project2poly.m


quadtree.m


quality.m


tinterp.m


refine.m

  REFINE: Refine triangular meshes.

 Quadtree triangle refinement is performed, with each triangle split into
 four sub-triangles. The new triangles are created by joining nodes
 introduced at the edge midpoints. The refinement is "quality" preserving,
 with the aspect ratio of the sub-triangles being equal to that of the
 parent.

  UNIFORM REFINEMENT:

  [p,t] = refine(p,t);

  p : Nx2 array of nodal XY coordinates, [x1,y1; x2,y2; etc]
  t : Mx3 array of triangles as indices, [n11,n12,n13; n21,n22,n23; etc]

  NON-UNIFORM REFINEMENT:

 Non-uniform refinement can also be performed by specifying which
 triangles are to be refined. Quadtree refinement is performed on
 specified triangles. Neighbouring triangles are also refined in order to
 preserve mesh compatibility. These triangles are refined using
 bi-section.

  [p,t] = refine(p,t,ti);

  ti : Mx1 logical array, with Ti(k) = TRUE if kth triangle is to be
       refined

 Functions defined on the nodes in P can also be refined using linear
 interpolation through an extra input:

  [p,t,f] = refine(p,t,ti,f);

  f : NxK array of nodal function values. Each column in F corresponds to
      a dependent function, F(:,1) = F1(P), F(:,2) = F2(P), etc.

 It is often useful to smooth the refined mesh using SMOOTHMESH. Generally
 this will improve element quality.

 Example:

   [p,t] = refine(p,t,ti);

 See also SMOOTHMESH, MESH2D


smoothmesh.m

  SMOOTHMESH: Smooth a triangular mesh using Laplacian smoothing.

 Laplacian smoothing is an iterative process that generally leads to an
 improvement in the quality of the elements in a triangular mesh.

  [p,t] = smoothmesh(p,t);

  p     : Nx2 array of nodal XY coordinates, [x1,y1; x2,y2; etc].
  t     : Mx3 array of triangles as indices, [n11,n12,n13; 
                                              n21,n22,n23; etc].
  maxit : Maximum allowable iterations.
  tol   : Convergence tolerance (Percentage change in edge length must be 
          less than TOL).

 If MAXIT or TOL are left empty the default values MAXIT = 20 and TOL =
 0.01 are used.

  EXAMPLE:

  [p,t] = smoothmesh(p,t,10,0.05);

 See also MESH2D, REFINE


C:\Users\AB\DArT_Toolshed\Algorithms\MeshGeneration\distmesh

boundedges.m

BOUNDEDGES Find boundary edges from triangular mesh
   E=BOUNDEDGES(P,T)


circumcenter.m

   Copyright (C) 2004-2005 Per-Olof Persson. See COPYRIGHT.TXT for details.


dcircle.m

   Copyright (C) 2004-2005 Per-Olof Persson. See COPYRIGHT.TXT for details.


ddiff.m

   Copyright (C) 2004-2005 Per-Olof Persson. See COPYRIGHT.TXT for details.


dexpr.m

   Copyright (C) 2004-2005 Per-Olof Persson. See COPYRIGHT.TXT for details.


dintersect.m

   Copyright (C) 2004-2005 Per-Olof Persson. See COPYRIGHT.TXT for details.


distmesh2d.m

DISTMESH2D 2-D Mesh Generator using Distance Functions.
   [P,T]=DISTMESH2D(FD,FH,H0,BBOX,PFIX,FPARAMS)

      P:         Node positions (Nx2)
      T:         Triangle indices (NTx3)
      FD:        Distance function d(x,y)
      FH:        Scaled edge length function h(x,y)
      H0:        Initial edge length
      BBOX:      Bounding box [xmin,ymin; xmax,ymax]
      PFIX:      Fixed node positions (NFIXx2)
      FPARAMS:   Additional parameters passed to FD and FH

   Example: (Uniform Mesh on Unit Circle)
      fd=inline('sqrt(sum(p.^2,2))-1','p');
      [p,t]=distmesh2d(fd,@huniform,0.2,[-1,-1;1,1],[]);

   Example: (Rectangle with circular hole, refined at circle boundary)
      fd=inline('ddiff(drectangle(p,-1,1,-1,1),dcircle(p,0,0,0.5))','p');
      fh=inline('min(4*sqrt(sum(p.^2,2))-1,2)','p');
      [p,t]=distmesh2d(fd,fh,0.05,[-1,-1;1,1],[-1,-1;-1,1;1,-1;1,1]);

   See also: MESHDEMO2D, DISTMESHND, DELAUNAYN, TRIMESH.


distmeshnd.m

DISTMESHND N-D Mesh Generator using Distance Functions.
   [P,T]=DISTMESHND(FDIST,FH,H,BOX,FIX,FDISTPARAMS)

      P:           Node positions (NxNDIM)
      T:           Triangle indices (NTx(NDIM+1))
      FDIST:       Distance function
      FH:          Edge length function
      H:           Smallest edge length
      BOX:         Bounding box [xmin,xmax;ymin,ymax; ...] (NDIMx2)
      FIX:         Fixed node positions (NFIXxNDIM)
      FDISTPARAMS: Additional parameters passed to FDIST

   Example: Unit ball
      dim=3;
      d=inline('sqrt(sum(p.^2,2))-1','p');
      [p,t]=distmeshnd(d,@huniform,0.2,[-ones(1,dim);ones(1,dim)],[]);

   See also: DISTMESH2D, DELAUNAYN, TRIMESH, MESHDEMOND.


dmatrix.m

   Copyright (C) 2004-2005 Per-Olof Persson. See COPYRIGHT.TXT for details.


dmatrix3d.m

   Copyright (C) 2004-2005 Per-Olof Persson. See COPYRIGHT.TXT for details.


dpoly.m

   Copyright (C) 2004-2005 Per-Olof Persson. See COPYRIGHT.TXT for details.


drectangle.m

   Copyright (C) 2004-2005 Per-Olof Persson. See COPYRIGHT.TXT for details.


drectangle0.m

   Copyright (C) 2004-2005 Per-Olof Persson. See COPYRIGHT.TXT for details.


dsphere.m

   Copyright (C) 2004-2005 Per-Olof Persson. See COPYRIGHT.TXT for details.


dunion.m

   Copyright (C) 2004-2005 Per-Olof Persson. See COPYRIGHT.TXT for details.


fixmesh.m

FIXMESH  Remove duplicated/unused nodes and fix element orientation.
   [P,T]=FIXMESH(P,T)


hmatrix.m

   Copyright (C) 2004-2005 Per-Olof Persson. See COPYRIGHT.TXT for details.


hmatrix3d.m

   Copyright (C) 2004-2005 Per-Olof Persson. See COPYRIGHT.TXT for details.


huniform.m

   Copyright (C) 2004-2005 Per-Olof Persson. See COPYRIGHT.TXT for details.


meshdemo2d.m

MESHDEMO2d Distmesh2d examples.


meshdemond.m

MESHDEMOND distmeshnd examples.


protate.m

   Copyright (C) 2004-2005 Per-Olof Persson. See COPYRIGHT.TXT for details.


pshift.m

   Copyright (C) 2004-2005 Per-Olof Persson. See COPYRIGHT.TXT for details.


simpplot.m

   Copyright (C) 2004-2005 Per-Olof Persson. See COPYRIGHT.TXT for details.


simpqual.m

SIMPQUAL Simplex quality.
   Q=SIMPQUAL(P,T,TYPE)

   TYPE == 1: Radius Ratio (default)
   TYPE == 2: Approximate


simpvol.m

SIMPVOL Simplex volume.
   V=SIMPVOL(P,T)


surftri.m

SURFTRI Find surface triangles from tetrahedra mesh
   TRI=SURFTRI(P,T)


uniformity.m

   Copyright (C) 2004-2005 Per-Olof Persson. See COPYRIGHT.TXT for details.


C:\Users\AB\DArT_Toolshed\Algorithms\MultiScaleCorrelatedLearning

C:\Users\AB\DArT_Toolshed\Algorithms\MultiScaleCorrelatedLearning\matlabfiles_aih

C:\Users\AB\DArT_Toolshed\Algorithms\MultiScaleCorrelatedLearning\matlabfiles_aih\@tree

dataFromLayer.m


display.m

DISPLAY Display array.
   DISPLAY(X) is called for the object X when the semicolon is not used
   to terminate a statement. 

   For example,
     X = inline('sin(x)')
   calls DISPLAY(X) while
     X = inline('sin(x)');
   does not.

   A typical implementation of DISPLAY calls DISP to do most of the work
   and looks as follows.  Note that DISP does not display empty arrays.

      function display(X)
      if isequal(get(0,'FormatSpacing'),'compact')
         disp([inputname(1) ' =']);
         disp(X);
      else
         disp(' ');
         disp([inputname(1) ' =']);
         disp(' ');
         disp(X);
      end
   
   See also INPUTNAME, DISP, EVALC.
   Overloaded methods:
      opaque/display
      sfit/display
      tree/display
      PointModelTemplate/display
      AAM_Element/display
      AAM/display
      avifile/display
      VideoReader/display
      cdfepoch/display
      inline/display
      timer/display
      serial/display
      ftp/display
      tscollection/display
      timeseries/display
      SimTimeseries/display
      phytree/display
      zpk/display
      tf/display
      ss/display
      realp/display
      pidstd/display
      pid/display
      loopswitch/display
      genss/display
      genmat/display
      genfrd/display
      daqdevice/display
      daqchild/display
      distributed/display
      codistributor2dbc/display
      codistributor1d/display
      codistributed/display
      gpuArray/display
      qfft/display
      vdspdebug/display
      eclipseide/display
      ccsrtdx/display
      ccsdebug/display
      imaqdevice/display
      imaqchild/display
      iviconfigurationstore/display
      icgroup/display
      instrument/display
      network/display
      frd/display
      cvtest/display
      cvdata/display
      dataset/display
      categorical/display
      vrworld/display
      vrnode/display
      vrfigure/display
      piecewisedistribution/display
      gmdistribution/display
      classregtree/display
      ProbDist/display
      NaiveBayes/display
      sym/display
   Reference page in Help browser
      doc display


drawtree.m

 function printtree(s)

 A simple method that shows the general structure of your tree

 Dr. A. I. Hanna (2007)
   Overloaded methods:
      tree/drawtree


get.m

GET    Get object properties.
   V = GET(H,'PropertyName') returns the value of the specified
   property for the graphics object with handle H.  If H is a 
   vector of handles, then get will return an M-by-1 cell array
   of values where M is equal to length(H).  If 'PropertyName' is
   replaced by a 1-by-N or N-by-1 cell array of strings containing
   property names, then GET will return an M-by-N cell array of
   values.

   GET(H) displays all property names and their current values for
   the graphics object with handle H.

   V = GET(H) where H is a scalar, returns a structure where each
   field name is the name of a property of H and each field contains
   the value of that property.

   V = GET(0, 'Factory') 
   V = GET(0, 'Factory<ObjectType>')
   V = GET(0, 'Factory<ObjectType><PropertyName>') 
   returns for all object types the factory values of all properties
   which have user-settable default values.  

   V = GET(H, 'Default') 
   V = GET(H, 'Default<ObjectType>') 
   V = GET(H, 'Default<ObjectType><PropertyName>') 
   returns information about default property values (H must be
   scalar).  'Default' returns a list of all default property values
   currently set on H.  'Default<ObjectType>' returns only the
   defaults for properties of <ObjectType> set on H.
   'Default<ObjectType><PropertyName>' returns the default value
   for the specific property, by searching the defaults set on H
   and its ancestors, until that default is found.  If no default
   value for this property has been set on H or any ancestor of H
   up through the root, then the factory value for that property
   is returned.
 
   Defaults can not be queried on a descendant of the object, or on the
   object itself - for example, a value for 'DefaultAxesColor' can not
   be queried on an axes or an axes child, but can be queried on a figure
   or on the root.

   When using the 'Factory' or 'Default' GET, if PropertyName is 
   omitted then the return value will take the form of a
   structure in which each field name is a property name and the 
   corresponding value is the value of that property.  If 
   PropertyName is specified then a matrix or string value will be
   returned.
   

   See also SET, RESET, DELETE, GCF, GCA, FIGURE, AXES.
   Overloaded methods:
      tree/get
      VolViewer/get
      PointModelTemplate/get
      AAM_Element/get
      AAM/get
      avifile/get
      scribehgobj/get
      scribehandle/get
      hgbin/get
      framerect/get
      figobj/get
      celltext/get
      cellline/get
      axistext/get
      axisobj/get
      axischild/get
      arrowline/get
      timer/get
      serial/get
      hgsetget/get
      RandStream/get
      COM/get
      tscollection/get
      timeseries/get
      phytree/get
      daqdevice/get
      daqchild/get
      qfft/get
      imaqdevice/get
      imaqchild/get
      iviconfigurationstore/get
      icgroup/get
      icdevice/get
      instrument/get
      InputOutputModel/get
      opcond.get
      dataset/get
      fdspec/get
      fdmeas/get
      fdline/get
      fdax/get
      vrworld/get
      vrnode/get
      vrfigure/get
   Reference page in Help browser
      doc get


get_max_levels.m


getchild.m


getchildren.m


getdata.m

--- help for daqchild/getdata ---
GETDATA Return acquired data samples from engine to MATLAB workspace.

    DATA = GETDATA(OBJ) returns the number of samples specified in the
    SamplesPerTrigger property of analog input object OBJ.  DATA is a
    M-by-N matrix where M is the number of samples returned and N is the
    number of channels in OBJ.  OBJ must be a 1-by-1 analog input object.

    DATA = GETDATA(OBJ, SAMPLES) returns the specified number, SAMPLES, 
    of data.
 
    [DATA, TIME] = GETDATA(OBJ) returns the number of samples specified 
    in the SamplesPerTrigger property of analog input object OBJ in 
    time-value pairs. TIME is a M-by-1 matrix where M is the number of 
    samples returned.

    [DATA, TIME] = GETDATA(OBJ,SAMPLES) returns the specified number, 
    SAMPLES, of data in time-value pairs.

    DATA = GETDATA(OBJ, SAMPLES, TYPE)
    [DATA, TIME] = GETDATA(OBJ, SAMPLES, TYPE) allows for DATA to be 
    returned as the data type specified by the string TYPE.  TYPE can either
    be 'double', for data to be returned as doubles, or 'native', for data 
    to be returned in its native data type.  If TYPE is not specified, 
    'double' is used as the default.

    [DATA, TIME, ABSTIME] = GETDATA(...) returns the absolute time ABSTIME  
    of the trigger, which can also be found in OBJ's InitialTriggerTime
    property.  ABSTIME is returned as a CLOCK vector.

    [DATA, TIME, ABSTIME, EVENTS] = GETDATA(...) returns the structure EVENTS 
    which contains a log of events associated with OBJ.

    GETDATA is a blocking function that returns execution control to the 
    MATLAB workspace once the requested number of samples become available. 
    OBJ's SamplesAvailable property will automatically be reduced by the 
    number of samples returned by GETDATA.  If the requested number of samples
    is greater than the samples to be acquired, then an error is returned.

    TIME=0 is defined as the point at which data logging begins, i.e., OBJ's 
    Logging property is set to 'On'.  TIME is measured continuously, in
    seconds, with respect to 0 until the acquisition is stopped, i.e., OBJ's
    Running property is set to 'Off'.

    If GETDATA returns data from multiple triggers, the data from each 
    trigger is separated by a NaN.  This will increase the length of DATA 
    and TIME by the number of triggers.  If multiple triggers occur, 
    ABSTIME, is the absolute time of the first trigger.

    It is possible to issue a ^C (Control-C) while GETDATA is blocking.  This
    will not stop the acquisition but will return control to MATLAB.

    See also DAQHELP, FLUSHDATA, GETSAMPLE, PEEKDATA, PROPINFO.

   Overloaded methods:
      analoginput/getdata
      imaqdevice/getdata


increment_level.m


istree.m


leafCoords.m


printtree.m

printtree( t, indent )
   Print a tree whose leaves are integers.
   Overloaded methods:
      tree/printtree


set.m

SET    Set object properties.
   SET(H,'PropertyName',PropertyValue) sets the value of the
   specified property for the graphics object with handle H.
   H can be a vector of handles, in which case SET sets the
   properties' values for all the objects.

   SET(H,a) where a is a structure whose field names are object
   property names, sets the properties named in each field name
   with the values contained in the structure.

   SET(H,pn,pv) sets the named properties specified in the cell array
   of strings pn to the corresponding values in the cell array pv for
   all objects specified in H.  The cell array pn must be 1-by-N, but
   the cell array pv can be M-by-N where M is equal to length(H) so 
   that each object will be updated with a different set of values
   for the list of property names contained in pn.

   SET(H,'PropertyName1',PropertyValue1,'PropertyName2',PropertyValue2,...)
   sets multiple property values with a single statement.  Note that it
   is permissible to use property/value string pairs, structures, and
   property/value cell array pairs in the same call to SET.

   A = SET(H, 'PropertyName') 
   SET(H,'PropertyName')
   returns or displays the possible values for the specified
   property of the object with handle H.  The returned array is
   a cell array of possible value strings or an empty cell array
   if the property does not have a finite set of possible string
   values.
   
   A = SET(H) 
   SET(H) 
   returns or displays all property names and their possible values for
   the object with handle H.  The return value is a structure whose
   field names are the property names of H, and whose values are 
   cell arrays of possible property values or empty cell arrays.

   The default value for an object property can be set on any of an 
   object's ancestors by setting the PropertyName formed by
   concatenating the string 'Default', the object type, and the 
   property name.  For example, to set the default color of text objects
   to red in the current figure window:

      set(gcf,'DefaultTextColor','red')
   
   Defaults can not be set on a descendant of the object, or on the
   object itself - for example, a value for 'DefaultAxesColor' can not
   be set on an axes or an axes child, but can be set on a figure or on
   the root.

   Three strings have special meaning for PropertyValues:
     'default' - use default value (from nearest ancestor)
     'factory' - use factory default value
     'remove'  - remove default value.

   See also GET, RESET, DELETE, GCF, GCA, FIGURE, AXES.
   Overloaded methods:
      BioSeq/set
      BioRead/set
      BioMap/set
      tree/set
      VolViewer/set
      PointModelTemplate/set
      AAM_Element/set
      AAM/set
      avifile/set
      scribehgobj/set
      scribehandle/set
      framerect/set
      figobj/set
      editrect/set
      editline/set
      celltext/set
      cellline/set
      axistext/set
      axisobj/set
      axischild/set
      arrowline/set
      timer/set
      serial/set
      hgsetget/set
      RandStream/set
      COM/set
      tscollection/set
      timeseries/set
      phytree/set
      daqdevice/set
      daqchild/set
      qfft/set
      imaqdevice/set
      imaqchild/set
      iviconfigurationstore/set
      icgroup/set
      icdevice/set
      instrument/set
      InputOutputModel/set
      DynamicSystem/set
      opcond.set
      dataset/set
      fdspec/set
      fdmeas/set
      fdline/set
      fdax/set
      vrworld/set
      vrnode/set
      vrfigure/set
   Reference page in Help browser
      doc set


set_max_levels.m


setchild.m

--- help for tree/setchild ---
 function T = setchild(T, indx, child)

 Dr. A. I. Hanna (2007)


setchildren.m


setdata.m


tree.m


C:\Users\AB\DArT_Toolshed\Algorithms\MultiScaleCorrelatedLearning\matlabfiles_aih\Data

C:\Users\AB\DArT_Toolshed\Algorithms\MultiScaleCorrelatedLearning\matlabfiles_aih\DataGenFunctions

cllib_blurimage.m

 function varargout = cllib_blurimage(varargin)

 Inputs
 'image' - the input image
 'H' - the blur mask (optional)
 'hsize' - the size of the blur mask (only used if 'H' is not supplied (default = 20)
 'sigma' - the std of the blur mask (only used if 'H' is not supplied (default = 10)


 Outputs
 'bim' - the blurred image

 Example:
 I = imread('pout.tif');
 bim = cllib_blurimage('image', I);
 subplot(1,2,1); imagesc(I); axis image;
 subplot(1,2,2); imagesc(bim); axis image;


 Example:
 I = imread('pout.tif');
 bim = cllib_blurimage('image', I, 'hsize', 10, 'sigma', 5);
 subplot(1,2,1); imagesc(I); axis image;
 subplot(1,2,2); imagesc(bim); axis image;


 Dr. A. I. Hanna (2007)


cllib_embedImage.m

 function varargout = cllib_embedImage(varargin)

 Inputs
 'image' - the image to embed
 'horiz_border' - the number of pixels padding each side of the image
                 (default = 200)
 'vert_border' - the number of pixels padding the top and bottom of the
                 image (default = 200)
 'fillval' - the fill value (default = 0)

 Dr. A. I. Hanna (2007)


cllib_genBlurredImages.m

 function blurred_images = cllib_genBlurredImages(WINDOW_TREE, im, N)

 Generates the blurred images needed by each level in the tree defined by
 WINDOW_TREE.

 Dr. A. I. Hanna 


cllib_genLevelImages.m

 function [varargout] = cllib_genLevelImages(varargin)

 Generates a series of images ready for PCA
 
 Inputs
 'image' - the original input image
 'N' - the number of samples to generate (default = 50)
 'range' - the range of jiggle (1x2 vector) (default = [50 50])
 'subsampsz' - the sub sample size (1x2 vector) (default = [16 16])
 'hsize' - the size of the blurring filter (default = 20)
 'theta' - the orientaion
 'sigma' - the standard deviation of the blurring filter (default = 10)

 Dr. A. I. Hanna (2007)


cllib_genSampleImage.m

 function [varargout] = cllib_genSampleImage(varargin)

 Generates an image ready for PCA
 
 Inputs
 'image' - the original input image
 'window' - the window to look through
 'range' - 1x2 range of translation of the image (default = [50 50])
 'subsampsz' - the sub sample size (1x2 vector) (default = [16 16])
 'hsize' - the size of the blurring filter (default = 20)
 'sigma' - the standard deviation of the blurring filter (default = 10)


 Dr. A. I. Hanna (2007)


cllib_genVirtualAppleScene.m


C:\Users\AB\DArT_Toolshed\Algorithms\MultiScaleCorrelatedLearning\matlabfiles_aih\DispFunctions

cllib_drawImage.m

 function varargout = cllib_drawImage(varargin)


 Inputs:
 'image' - the image to draw
 'text' - the title of the image
 'parent' - the parent to draw to (default = gca)

 Dr. A. I. Hanna (2007)


cllib_drawWindow.m

 function varargout = cllib_drawWindow(varargin)

 Initializes a window for training multi-scale hierarchical piecewise
 linear statistical models.

 Inputs:
 'window' - the window to draw
 'parent' - the parent to draw to (default = gca)
 'schema' - the type of plot (0 = default colour,
            1 = black window, 2 = dot at centroid)

 Dr. A. I. Hanna (2007)


cllib_moveImagesInStack.m

 function cllib_moveImagesInStack(im_handles, offset, xdata_orig, ydata_orig)

 Takes the handle to a set of images in a tree and moves them to the
 position given in offset.


cllib_showImageStack.m

 function varargout = cllib_showImageStack(varargin)


 Dr. A. I. Hanna (2008)


cllib_show_result.m

 function [varargout] = cllib_show_result(varargin)



 Dr. A. I. Hanna (2007)


cllib_show_schema.m

 function [varargout] = cllib_show_schema(varargin)



 Dr. A. I. Hanna (2007)


cllib_threeImgFigure.m


drawEdge.m


drawnode.m


drawtree.m

 function printtree(s)

 A simple method that shows the general structure of your tree

 Dr. A. I. Hanna (2007)
   Overloaded methods:
      tree/drawtree


plotParameterModel.m

 function varargout = genParameterModel(varargin)


 Dr. A. I. Hanna (2007)


C:\Users\AB\DArT_Toolshed\Algorithms\MultiScaleCorrelatedLearning\matlabfiles_aih\Hebbian

generalized_hebbian.m

 function varargout = generalized_hebbian(varargin)
 
 This is the generalized hebbian algorithm (GHA) as given by Haykin. In
 this algorithm we are performing a Gram-Schmidt ish approach to
 processing the data. In other words we are removing the component of the
 weights from the data in turn.

 "Neural Networks, A Comprehensive Foundation", 2nd ed, S. Haykin, page 414

 Inputs:
  'X' - the input data
  'W' - the weight matrix
  'eta' - the learning rate

 Outputs:
  'Y' - the output from the network
  'W' - the updated weight matrix

 Example:

 generalized_hebbian('demo', 1);
 
 Dr. A. I. Hanna (2007)


hebbian.m

 function varargout = hebbian(varargin)

 A Matlab function that performs Hebbian learning on a data set.


C:\Users\AB\DArT_Toolshed\Algorithms\MultiScaleCorrelatedLearning\matlabfiles_aih\IOFunctions

cllib_loadImage.m

 function varargout = cllib_loadImage(varargin)


 Inputs:
 'filename' - the name of the image to load
 'scale' - the scale of the image to return (default = 1)

 Outputs:
 'image' - the image

 Dr. A. I. Hanna (2007)


cllib_loadImagesFromDir.m

 function images = cllib_loadImagesFromDir(varargin)

 A method that loads a series of images from disk. 

 Input Parametes:

 'scale' - the scaling factor of the image i.e. scale = 0.5 would half the
           size of the image. (Images should be approx 512x512)

 Example 1:

 images = cllib_loadImagesFromDir;

 Example 2:

 images = cllib_loadImagesFromDir('scale', 0.5);



 Dr. A. I. Hanna, CMP, UEA, 2007.


cllib_loadResult.m

 function varargout = cllib_loadResult(varargin)


 Inputs:
 'filename' - the name of the image to load

 Outputs:
 'model' - the model

 Dr. A. I. Hanna (2007)


cllib_saveTestResult.m

 function cllib_saveTestResult(varargin)

 % model is created by cllib_test

 cllib_saveTestResult('model', model);

 Dr. A. I. Hanna, CMP, UEA, 2007.


C:\Users\AB\DArT_Toolshed\Algorithms\MultiScaleCorrelatedLearning\matlabfiles_aih\ModelFunctions

cllib_applyLevelResponse.m

 function model = cllib_applyLevelResponse(model, level_response, level_id)


 Dr. A. I. Hanna (2007)


cllib_applyResidualResponse.m

 function model = cllib_applyResidualResponse(model, level_response, level_id)

 Dr. A. I. Hanna, CMP, UEA, 2007.


cllib_applyResponse.m


cllib_buildTree.m

 function [varargout] = cllib_buildTree(varargin)

 Builds an empty tree ready for population by other methods in cllib.


 Dr. A. I. Hanna (2007)


cllib_build_shapp_model.m

 function cllib_build_shapp_model

 Builds a shapperance model.

 1) Prompts user for the filename where the model should be saved on disk.
 2) Asks the user to select result files generated by track_image_series.
 3) To walk through the shappearance model use cllib_walk_shapp_model

 Example

 cllib_build_shapp_mod

 Dr. A. I. Hanna, CMP, UEA, 2007.


cllib_build_window_shape_model.m

 function [window_app_model] = cllib_build_window_shape_model(images)

 Takes a cell array of results where each

 Dr. A. I. Hanna, CMP, 2008.


cllib_calcAppModel.m

 function [varargout] = cllib_calcAppModel(varargin)


 Inputs:
 'data' - a NxM matrix of input data where each column is an observation
 'pc_ind' - the 1xN vector of PC indices used in the model (default = [1 2])


 Dr. A. I. Hanna (2007)


cllib_createModel.m

 function varargout = cllib_createModel(varargin)

 Creates and saves a model for a particular image. 
 
 Usage:
 cllib_createModel;

 Dr. A. I. Hanna, CMP & JIC, 2008.


cllib_estimateLevelResponses.m

 function [model, level_est] = cllib_estimateLevelResponses(model, level_id, test_image, model_id)

 Dr. A. I. Hanna, CMP & JIC, 2008.


cllib_estimateModelResponses.m

 function [model] = cllib_estimateModelResponses(model, level_id, test_image)


 Dr. A. I. Hanna, CMP, UEA, Norwich, UK, 2007.


cllib_estimateWindowResponse.m

 function [window] = cllib_estimateWindowResponse(img, window, model_id, levelmodel)

 Takes an image and a window and calculates the response given a
 particular model.

 Dr. A. I. Hanna (2007)


cllib_genParameterModel.m

 function varargout = genParameterModel(varargin)


 Dr. A. I. Hanna (2007)


cllib_makeParamModel.m


cllib_mergeModels.m


cllib_param2response.m


cllib_paramInference.m


cllib_test.m

 function [varargout] = cllib_test(varargin)

 Dr. A. I. Hanna (2007)


cllib_train.m

 function [varargout] = cllib_train(varargin)

 Builds a tree structure at all levels by finding correlations between
 appreance models and response amplitudes.


 Dr. A. I. Hanna (2007)


cllib_trainlevel.m

 function [varargout] = cllib_trainlevel(varargin)

 Trains a particular level of the tree using a specified image.

 Dr. A. I. Hanna (2007)


cllib_tree2pointmodel.m

 function  [pm, id] = cllib_tree2pointmodel(T, level, pm, id)

 Given a tree it returns a matrix where each row represents the centroid
 of window.

 Dr. A. I. Hanna (2007)


cllib_walk_shapp_model.m

 function cllib_walk_shapp_model

 Walks along a shappearance model that you select.

 1) Prompts the user to open a shappearance model and displays a walk
 along the first 3 pcs from -3 SD and +3 SD.

 Example:

 cllib_walk_shapp_model

 Dr. A. I. Hanna, CMP, UEA, 2007.


C:\Users\AB\DArT_Toolshed\Algorithms\MultiScaleCorrelatedLearning\matlabfiles_aih\Results

C:\Users\AB\DArT_Toolshed\Algorithms\MultiScaleCorrelatedLearning\matlabfiles_aih\WindowFunctions

cllib_addWindowToTree.m


cllib_addWindowsToTree.m


cllib_calcRectOverlap.m


cllib_dispWindowImage.m

 function varargout = cllib_dispWindowImage(varargin)

 'window'
 'level_id'
 'window_im'

 Dr. A. I. Hanna (2007)


cllib_drawInformativeWindows.m

 function T = cllib_drawInformativeWindows(T, level)

 Draws all the informative windows for a given tree.

 Dr. A. I. Hanna (2007)


cllib_getAllWindows.m


cllib_getInformativeWindows.m

 function infwins = cllib_getInformativeWindows(T, infwins, level)


 Dr. A. I. Hanna (2007)


cllib_getWindowRect.m


cllib_image2windowview.m

 function varargout = cllib_image2windowview(varargin)

 'image'
 'window'

 Dr. A. I. Hanna (2007)


cllib_initWindow.m

 function window = cllib_initWindow(varargin)

 Initializes a window for training multi-scale hierarchical piecewise
 linear statistical models.

 Inputs
 'window_size' - the size of the window (1x2 vector) (default = [100 100])
 'color' - the colour of the window (1x3 vector) (default = [1 0 0])
 'sub_samp_size' - the sub sample size of this window to apply to images (default = [32 32])
 'mu' - the centroid of the window (1x2 vector) (default = [0 0])
 'mean_response' - mean response of the window, i.e. movement x, y, z (default = [0 0])

 See also: splitWindow, translateWindow

 Dr. A. I. Hanna (2007)


cllib_moveTree.m

 function T = cllib_moveTree(T, pos)

 Moves the tree to position pos

 Dr. A. I. Hanna (2007)


cllib_setInformativeWindows.m


cllib_setWindowPositions.m

T = model.WINDOW_TREE;


cllib_splitWindow.m

 function varargout = cllib_splitWindow(varargin)

 Splits a window into quarters and returns each of the subwindows 

 Inputs
 'window' - the window to be split

 Outputs
 'subwindows' - a cell array containing the four sub windows.%

 See also: initWindow, translateWindow

 Dr. A. I. Hanna (2007)


cllib_splitWindow1d.m

 function varargout = cllib_splitWindow1d(varargin)

 Splits a window into quarters and returns each of the subwindows 

 Inputs
 'window' - the window to be split

 Outputs
 'subwindows' - a cell array containing the four sub windows.%

 See also: initWindow, translateWindow

 Dr. A. I. Hanna (2007)


cllib_splitWindow2d.m

 function varargout = splitWindow2d(varargin)

 Splits a window into quarters and returns each of the subwindows 

 Inputs
 'window' - the window to be split

 Outputs
 'subwindows' - a cell array containing the four sub windows.%

 See also: initWindow, translateWindow

 Dr. A. I. Hanna (2007)


cllib_translateChildren.m

 function T = cllib_translateChildren(T, level, trans)

 Translates all the children from T down by trans

 Dr. A. I. Hanna (2007)


cllib_translateWindow.m

 function varargout = cllib_translateWindow(varargin)

 Translates a window in the xy plane 

 Inputs
 'window' - the window to be translated
 'offset' - the offset (1x2 vector) (default = [0 0])

 Outputs
 'window' - the translated window

 See also: initWindow, splitWindow

 Dr. A. I. Hanna (2007)


cllib_updateChildrenInf.m


cllib_windowOverlap.m


cllib_zeroAllWindows.m


C:\Users\AB\DArT_Toolshed\Algorithms\Segmentation

C:\Users\AB\DArT_Toolshed\Algorithms\Segmentation\LevelSets

active_contour_chan_vese2d.m

 function varargout = active_contour_chan_vese2d(varargin)

 A method that implements the Chan-Vese model applied to image
 segmentation. This model is a special case of the Mumford-Shah model and
 it tries to separate the image into regions based on intensities.

 Default parameters:

 'image' - the input image (default = 'pout.tif')
 'phi' - the initial distance function (default = circular disk)
 'lambda1' - weight of energy inside the contour (default = 0.5)
 'lambda2' - weight of energy outside the contour (default = 0.5)
 'mu' - weighting for the length of the contour (default = 1)
 'nu' - weighting of area of the contour (default = 1)
 'deltaT' - step size for the update of phi (default = 0.5, note: 0.01<=deltaT<=0.9)
 'verbose' - shows the evolution of the curve phi (default = 0)
 'max_iter' - the maximum number of iterations to perform (default = 10)
 'normforceflag' - normalizes the force term at every iteration (default = 1)

 Example:

 active_contour_chan_vese2d('verbose', 1);

 Paper: "Minimizing Functionals: A Level Set Approach", A. I. Hanna, School of Computing Sciences, University of East Anglia, 2007.
 URL: http://www2.cmp.uea.ac.uk/~aih/papers/MinimizingFunctionalsUsingLevelSets/MinimizingFunctionals-ALevelSetApproach_hanna.pdf

 Dr. A. I. Hanna (2007)


active_contour_chan_vese3d.m

 function varargout = active_contour_chan_vese3d(varargin)

 A method that implements the Chan-Vese model applied to volume
 segmentation. This model is a special case of the Mumford-Shah model and
 it tries to separate the volume into sub-volumes based on intensities.

 Default parameters:

 'volume' - the input volume 
 'phi' - the initial distance function (default = circular disk)
 'lambda1' - weight of energy inside the contour (default = 0.5)
 'lambda2' - weight of energy outside the contour (default = 0.5)
 'mu' - weighting for the length of the contour (default = 1)
 'nu' - weighting of area of the contour (default = 1)
 'deltaT' - step size for the update of phi (default = 0.5, note: 0.01<=deltaT<=0.9)
 'verbose' - shows the evolution of the curve phi (default = 0)
 'max_iter' - the maximum number of iterations to perform (default = 10)
 'normforceflag' - normalizes the force term at every iteration (default = 1)

 Example:

 active_contour_chan_vese3d('verbose', 1);

 Paper: "Minimizing Functionals: A Level Set Approach", A. I. Hanna, School of Computing Sciences, University of East Anglia, 2007.
 URL: http://www2.cmp.uea.ac.uk/~aih/papers/MinimizingFunctionalsUsingLevelSets/MinimizingFunctionals-ALevelSetApproach_hanna.pdf

 Dr. A. I. Hanna (2007)


C:\Users\AB\DArT_Toolshed\Algorithms\Segmentation\LevelSets\images

lsShowResult.m

phi = load('../Results/leaf_phi_128_128_128.mat');
phi = phi.phi;
phi = smooth3(phi);


lsdemo_2d.m

 function lsdemo_2d(im_opt)

 Usage:
 lsdemo_2d(opt_id); % opt_id = 1, 2, 3, 4

 Dr. A. I. Hanna (2007)


lsdemo_3d.m


lsdispVolume.m

 function dispVolume(cdata, axish)

 A very very simple script to display a set of volume data

 Dr. A. I. Hanna (2006


C:\Users\AB\DArT_Toolshed\Algorithms\Segmentation\LevelSets\volumes

C:\Users\AB\DArT_Toolshed\Algorithms\Segmentation\LevelSets\volumes\ConfocalGFP

C:\Users\AB\DArT_Toolshed\Algorithms\Segmentation\LevelSets\volumes\Leaf

alglib_flo_spots.m

 function [NUM, c, L] = alglib_flo_spots(I)

 Segments out flo spots in images

 Dr. A. I. Hanna (2008)


alglib_white_spots.m

 function [NUM, c, L] = alglib_white_spots(I, I2)

 Segments out white spots in images

 Dr. A. I. Hanna (2008)


cells_SEM.m

%%%%%%%%%%%%%%%%%%%%

 Process the cell image

%%%%%%%%%%%%%%%%%%%%%


segment_DAPI.m

 function [bim] = segment_DAPI(varargin)

 A script used to segment DAPI images. Typically these images are tif
 images with a single channel.

 Dr. A. I. Hanna, CMP, 2008.


C:\Users\AB\DArT_Toolshed\Algorithms\Sieve

C:\Users\AB\DArT_Toolshed\Algorithms\Sieve\32-bit

runme.m

RUNME Summary of this function goes here
   Detailed explanation goes here


C:\Users\AB\DArT_Toolshed\Algorithms\Sieve\64-bit

SIEVE4.M

 function [Y1 Y2 ... Yn] = sieve(X, options, outputs, scan, type, time)

 Apply recursive median datasieve to matrix X with complete control of 
 polarity (sign of recorded granules or edges) and bias (sign of filtered
 extrema) at each mesh.  If X is a matrix then the columns of X are sieved 
 separately.  The scan argument controls the orientation of the sieve for
 angles to the vertical in the range 0 to 45 degrees (all other angles can
 be obtained by transposing or flipping the input data matrix).

 The OPTIONS matrix is of size m*4 and each row contains the following data:

	[maxmesh polarity bias output]

   Thus meshes up to maxmesh are processed with the specified polarity and
   bias parameters, and the results are stored according to the output
   parameter.  The polarity and bias parameters are set using:

	 0  : Bipolar operation.
	-1  : Process negative events.
 	+1  : Process positive events.

   Note that polarity constraints are a subset of bias constraints - a polarity
   of -1 means that only negative granules are recorded (but both positive and
   negative extrema are processed), while a bias of -1 means that only negative
   extrema are processed.

   The output parameter specifies the number of the output argument (1 to n)
   or 0 for no output.  The setting of the corresponding value in the OUTPUTS
   vector determines the form of the output.

 The OUTPUTS vector if a string containing the symbols 'l','b','g','e','s'.  
   These are concatenated into a string, for example 'lbbe'.  The k'th element 
   in the string corresponds to the k'th output argument, Yk.  The meaning of 
   the symbols is as follows:

	'l' : Lowpass   - Lowpass image at output of mesh maxmesh
	'b' : Bandpass  - Bandpass image from granules at processed meshes.
	'g' : Granules  - Granules list from processed meshes.
	'e' : Edge map  - Edge map formed from granule edges and amplitudes.
	's' : Sobel map - Edgemap with post processing to emulate sobel filter. 

 The SCAN parameter is a two element vector of integers that determines the
   orientation of the sieving.  The angle at which the sieve is applied 
   relative to the vertical is given in terms of the input vector [k1 k2] as
   atan((k1+k2)/2).  So for k1=k2=1, the scan is at 45 degs. For k1=1, k2=2, 
   the scan is at 34 degs.  For k1, k2 > image dimension the scan is vertical.
   Note that k1, k2 must be positive integers (so not all angles are possible).
   Normally, k1 and k2 should be equal or differ by 1.

   Note that the scan lines to which sieves are applied are guaranteed to be
   non-overlapping.  However, this means that the gap between samples on a 45
   degree scan is sqrt(2) times the normal sample spacing, which should be
   taken into account when specifying 'cut-off' meshes etc. 
  
 The TYPE parameter specifies whether the data is integer or double precision.
   It is a string that can be set to 'int' or 'double'.  If 'int' is specified,
   then the user is responsible for ensuring that the data range does not
   exceed a 32 bit signed integer.  If the data is not integer than it is 
   rounded using a 'C' cast. Integer precision is faster than double precision.
   Double is default.

 The TIME parameter is set to 1 to display elapsed user time while performing
   processing.  It excludes memory allocation overheads.  The default is 0.


 EXAMPLE:

   [Y1 Y2 Y3] = sieve(X, OPTIONS, OUTPUTS, [1 1])

   with OPTIONS = [ 4  0  0  1   and OUTPUTS = 'lbb'
                   16 +1  0  2
		    32  0  0  0
		    64  0  0  3]

   returns Y1 = low pass image after initial noise reduction up to mesh 4.
           Y2 = bandpass image containing positive granules for meshes 5 to 16.
           Y3 = bandpass image containing all granules for meshes 33 to 64.

   using scan lines at 45 degrees to vertical.


 Robert Young   13 October 1994


sieve2d.m

 FILE
	sieve2d.m

 AUTHOR
	rwy Cambridge Consultants (project leader J.A. Bangham, UEA)
 
 MODIFICATION HISTORY

	1.1	02:jun:95	rwy	First version submitted to SCCS.
	1.2	26:jun:95	rwy	Extended for open, close, M & N sieves.
	1.3	27:jul:95	rwy	Allows bandpass outputs & granule fusing
 	1.4	03:aug:95	rwy	Minor revision.
	1.5	09:aug:95	rwy	Added granule output.
	1.6	14:sep:95	rwy	Added edge map output.
   2.0 01/01/2013  jab Updated to 64 bit and includes additional outputs
	
 SCCS IDENTIFIER
	@(#)sieve2d.m	1.6 9/14/95
	@(#)SIV2d.m	2.0 01/01/2013


 [Y1 Y2 ... Ym] = sieve2d(X, [M1 M2 ... Mn], 'F1F2 ... Fn', 
						[O1 O2 ... On], type, N)

 Perform a 2-D connected-set datasieve based on area.

 X	:	input image
 Mk	:	maximum mesh for each filter
 Fk	:	filter type  (m, o, c, M or N)	[default: m]
 Ok	: 	output index for each filter    [default: 1 for last filter]
 type  :	output type  (l, b, e or f)	[default: l]
 N	:	connectivity (4 or 8)           [default: 4]
 Yj	:	output image from a filter 

 The input image is processed using one or more filters Fk, applied 
 sequentially, up to the associated maximum meshes Mk.  The available 
 filters are:

 m     :	median     (maxima & minima processed in arbitrary order)
 o     : 	opening    (only maxima processed)
 c     : 	closing    (only minima processed)
 M     : 	open-close (maxima then minima processed at each scale)
 N     : 	close-open (minima then maxima processed at each scale)

 For each filter, the result may optionally be output by specifying the
 index of the desired output argument as the input Ok (where Ok=1 represents 
 the first output argument).  If Ok=0 then no output is generated 
 for filter Fk.

 The available output types are:

 l	:	low-pass
 b	:	band-pass 
 f	:	band-pass with granule fusing
 e  	:	band-pass, returning edge map
 g	:	granule list

 where the band-pass output is the difference between the outputs of the
 current filter and the previous filter (or the input image) in the 
 sequence of filters.  The same output type must be used for each filter.

 The granule output list format is a vector containing the following data:

 	[image_vdim image_hdim gran1_len gran1_val gran1_pos gran2_len ...]

 where gran?_pos is a list of the positions of the granule pels within
 the image (one scalar index per pel).

 Note that producing multiple low or band pass outputs in this way is much 
 more efficient than calling the sieve function several times with different
 meshes.  

 Memory requirements are about 18 bytes per image pel.


 EXAMPLES:

 Filter up to mesh 100 using a median sieve, and generate a low pass output:

 	Y = sieve2d(x, 100)
 or 	Y = sieve2d(x, 100, 'm')
 or	Y = sieve2d(x, 100, 'm', 1)
 or	Y = sieve2d(x, 100, 'm', 1, 'l')

 Filter up to mesh 10 using a median sieve, then up to mesh 100 using an 
 opening operator, and output the band pass signal between meshes 11 and 100:

	Y = sieve2d(x, [10 100], 'mo', [0 1], 'b')

 Form a sequence of low pass outputs using closing operators:

	[Y1 Y2 Y3 Y4] = sieve2d(x, [8 16 32 64], 'cccc', [1 2 3 4])

 As above, with initial noise reduction using a median filter:

	[Y1 Y2 Y3 Y4] = sieve2d(x, [4 8 16 32 64], 'mcccc', [0 1 2 3 4])

 As above, but producing band pass outputs with granule fusing:

	[Y1 Y2 Y3 Y4] = sieve2d(x, [4 8 16 32 64], 'mcccc', [0 1 2 3 4], 'f')


sieve3d.m

 FILE
	sieve3d.m

 AUTHOR
	rwy (JAB)

 MODIFICATION HISTORY

 	1.3 Dec:1010    JAB Exploit 64 bit by extending int to __int64
 					It does handle 512x512x512 images but needs >8Gbytes memory
 					to avoid being very slow
 
 					input image is uint8
                   outputs are all double

	1.1	03:aug:95	rwy	jab First version submitted to SCCS.
	1.2	04:aug:95	rwy	Added depth resolution.
	
 SCCS IDENTIFIER
	@(#)sieve3d.m	1.2 8/4/95


 [Y1 Y2 ... Ym] = sieve3d(X, [depth Dres], [M1 M2 ... Mn], 'F1F2 ... Fn', 
						[O1 O2 ... On], type, N)

 Perform a 3-D connected-set datasieve based on volume.

 X	:	input image
 depth	:	image depth  (scalar)
 Dres	:	resolution in depth dimension 	[default: 1]
 Mk	:	maximum mesh for each filter
 Fk	:	filter type  (m, o, c, M or N)	[default: m]
 Ok	: 	output index for each filter    [default: 1 for last filter]
 type  :	output type  (l, b or f)	[default: l]
 N	:	connectivity (only 6 currently) [default: 6]
 Yj	:	output image from a filter 

 The input 3D image is supplied as a 2D matrix, X, in which the successive
 2D slices at each depth are appended to the input matrix as new columns.  
 So if the matrices for the slices are S1, S2, S3 ... Sd then:

 	X = [S1 S2 S3 ... Sd]	and   depth = d

 Currently, 3D volumes are 6-connected using the four 4-connected pels in the 
 same slice, and the pel in the same location within each of the two adjacent
 slices. 

 The resolution in the depth dimension can be specified relative to the
 resolution within each slice.  This is achieved by setting the Dres input 
 parameter to an integer greater than 1, in order to simulate a depth 
 resolution that is coarser by a factor Dres.  The 'volume' of each region
 is increased by adding an extra Dres-1 units for each connection that 
 exists between pels on different slices.  Thus, a region of N pels all in the 
 same slice has a volume N, while a line of N pels all on different slices
 has a volume N + (N-1)*(Dres-1). 

 The input image is processed using one or more filters Fk, applied 
 sequentially, up to the associated maximum meshes Mk.  The available 
 filters are:

 m     :	median     (maxima & minima processed in arbitrary order)
 o     : 	opening    (only maxima processed)
 c     : 	closing    (only minima processed)
 M     : 	open-close (maxima then minima processed at each scale)
 N     : 	close-open (minima then maxima processed at each scale)

 For each filter, the result may optionally be output by specifying the
 index of the desired output argument as the input Ok (where Ok=1 represents 
 the first output argument).  If Ok=0 then no output is generated 
 for filter Fk.

 The available output types are:

 l	:	low-pass
 b	:	band-pass 
 f	:	band-pass with granule fusing

 where the band-pass output is the difference between the outputs of the
 current filter and the previous filter (or the input image) in the 
 sequence of filters.  The same output type must be used for each filter.

 Note that producing multiple low or band pass outputs in this way is much 
 more efficient than calling the sieve function several times with different
 meshes. 

 Memory requirements are about 18 bytes per image pel.


 EXAMPLES:

 Filter an image composed of 50 slices up to mesh 100 using a median sieve, 
 and generate a low pass output:

 	Y = sieve3d(x, 50, 100)
 or 	Y = sieve3d(x, 50, 100, 'm')
 or	Y = sieve3d(x, 50, 100, 'm', 1)
 or	Y = sieve3d(x, 50, 100, 'm', 1, 'l')
 or	Y = sieve3d(x, [50 1], 100, 'm', 1, 'l')

 Filter up to mesh 10 using a median sieve, then up to mesh 100 using an 
 opening operator, and output the band pass signal between meshes 11 and 100.
 The resolution in the depth dimension is 3 times coarser than the
 resolution
 within each slice:

	Y = sieve3d(x, [50 3], [10 100], 'mo', [0 1], 'b')

 Form a sequence of low pass outputs using closing operators:

	[Y1 Y2 Y3 Y4] = sieve3d(x, 50, [8 16 32 64], 'cccc', [1 2 3 4])

 As above, with initial noise reduction using a median filter:

	[Y1 Y2 Y3 Y4] = sieve3d(x, 50, [4 8 16 32 64], 'mcccc', [0 1 2 3 4])

 As above, but producing band pass outputs with granule fusing:

	[Y1 Y2 Y3 Y4] = sieve3d(x, 50, [4 8 16 32 64], 'mcccc', [0 1 2 3 4],'f')


Bags.m


C:\Users\AB\DArT_Toolshed\Algorithms\Sieve\CPP

C:\Users\AB\DArT_Toolshed\Algorithms\Sieve\UpdatedSieve

ArrowRegions.m


ConnectedSets.m


MakeQuantiserTable.m


QuantiseDensity.m


C:\Users\AB\DArT_Toolshed\Algorithms\Sieve\UpdatedSieve\SIV_TestC_20130410

C:\Users\AB\DArT_Toolshed\Algorithms\Sieve\UpdatedSieve\SIV_TestC_20130410\directory_of_images_for_SIV

C:\Users\AB\DArT_Toolshed\Algorithms\Sieve\UpdatedSieve\SIV_TestD_20130410

C:\Users\AB\DArT_Toolshed\Algorithms\Sieve\UpdatedSieve\SIV_TestD_20130410\directory_of_images_for_SIV

C:\Users\AB\DArT_Toolshed\Algorithms\Sieve\UpdatedSieve\SIV_testA_20130410

C:\Users\AB\DArT_Toolshed\Algorithms\Sieve\UpdatedSieve\SIV_testB_20130410

C:\Users\AB\DArT_Toolshed\Algorithms\Sieve\UpdatedSieve\SIV_testB_20130410\directory_of_images_for_SIV

C:\Users\AB\DArT_Toolshed\Algorithms\Sieve\UpdatedSieve\SIV_test_20130410

SieveCTScans.m

 Program name: SieveCTScans
 Authors: GDT + PS
 Date:     15/09/2011


C:\Users\AB\DArT_Toolshed\Algorithms\Sieve\UpdatedSieve\Snapshots

checkbox_isometric.m


C:\Users\AB\DArT_Toolshed\Algorithms\Sieve\UpdatedSieve\directory_of_images_for_SIV

displayRegion_Graph_Vertex.m


display_Region_Graph_Vertex.m


extremalMSERUtility2.m

function [ imOut, granules, regions ] = extremalMSERUtility2( ( N ,siz, positions,diameters,graded) )

 To be compatible with ConnectedSets.m (i.e. the plotting axis)

data.showText

 Display a tree representation of a greyscale im


extremalTree.m

function [ imOut, granules, regions ] = extremalTree( ( N ,siz, positions,diameters,graded) )

 Display a tree representation of a greyscale im


extremalTreeFig.m


extremalTreeUtility.m

function [ imOut, granules, regions ] = extremalTreeUtility( ( N ,siz, positions,diameters,graded) )

data.showText

 Display a tree representation of a greyscale im


extremalTreeUtility2 - Copy.m

function [ imOut, granules, regions ] = extremalTreeUtility2( ( N ,siz, positions,diameters,graded) )

 To be compatible with ConnectedSets.m (i.e. the plotting axis)

data.showText

 Display a tree representation of a greyscale im


extremalTreeUtility2.m

function [ imOut, granules, regions ] = extremalTreeUtility2( ( N ,siz, positions,diameters,graded) )

 To be compatible with ConnectedSets.m (i.e. the plotting axis)

data.showText

 Display a tree representation of a greyscale im


quantise_uint8_im.m


saveCurrentStateSIV.m


showGranules.m

function showGranules
  
Toggle views of significant granules


showTheText.m


C:\Users\AB\DArT_Toolshed\Algorithms\Sieve\UpdatedSieve\sieveCode

sieveSect.m

 Program name: SIEVESECT
 Authors: GDT + PS
 Date:     15/09/2011


siv2d_m.m

 [Y1 Y2 ... Ym] = siv2d_m(X, [M1 M2 ... Mn], 'F1F2 ... Fn',
						[O1 O2 ... On], type, N, Levels)

 CALLS the mex code SIV2d.* To get a sequence of outputs call SIVE2d directly

 Perform a 2-D connected-set datasieve based on area.

 X	    :	uint8, input image
 Mk	:	maximum mesh (scale, area of extrema) for each filter
 Fk	:	filter type  (m, o, c, M or N)	[default: m]
 Ok	: 	output index for each filter    [default: 1 for last filter]
 type  :	type of output  (l, b, e, g, f or v)	[default: l]
 N	    :	connectivity (4 or 8)           [default: 4]
 EP    :   parameters for selecting extrema [Levels,Ratio]
              Levels :   number of finite levels in image [default: 256]
              Ratio  :   typically 5 [default: 1]
              MinArea:   typically 4 [default: 1]
              MaxArea:   typically 1000 [default: 1000]
              MinAmp :   typically 4 [default: 0]

 Yj	:	output image(j) from a filter

 FILTERS
 The input image is processed using one or more filters Fk, applied
 sequentially, up to the associated maximum meshes Mk.  The available
 filters are:

 m     :	median     (maxima & minima processed in arbitrary order)
 o     : 	opening    (only maxima processed, c.f. maximally stable extreme regions )
 c     : 	closing    (only minima processed, c.f. maximally stable extreme regions)
 M     : 	open-close (maxima then minima processed at each scale)
 N     : 	close-open (minima then maxima processed at each scale)

 For each filter, the result may optionally be output by specifying the
 index of the desired output argument as the input Ok (where Ok=1 represents
 the first output argument).  If Ok=0 then no output is generated
 for filter Fk.

 TYPE OF OUTPUT
 The available output types are:

 l	:	low-pass
 b	:	band-pass
 f	:	band-pass with granule fusing
 e :	band-pass, returning edge map
 g	:	granule list
 v :   verbose, i.e. main data structure of engine

 where the band-pass output is the difference between the outputs of the
 current filter and the previous filter (or the input image) in the
 sequence of filters.  The same output type must be used for each filter.

 The granule output list format is a cell array containing the following data:
 For example:
            Number: 3 (number of granules)
              area: [1 5 9]
             value: [18 7 11]
             level: [255 237 230]
         deltaArea: [4 4 4]
         last_area: [400 80 44]
              root: [76 76 76]
    PictureElement: {[77]  [77 94 60 78 76]  [77 95 61 93 59 94 60 78 76]}

 e.g.  data.g.PictureElement{3} contains indexes to the pixels that formed the granule

 Note that producing multiple low or band pass outputs in this way is much
 more efficient than calling the sieve function several times with different
 meshes.

 Memory requirements are about 18 bytes per image pel.


 EXAMPLES:

 Filter up to mesh 100 using a median sieve, and generate a low pass output:

 	Y = SIV2d(x, 100)
 or 	Y = SIV2d(x, 100, 'm')
 or	Y = SIV2d(x, 100, 'm', 1)
 or	Y = SIV2d(x, 100, 'm', 1, 'l')
 or	Y = SIV2d(x, 100, 'm', 1, 'l',4)
 or	Y = SIV2d(x, 100, 'm', 1, 'l',4,[256,4,5])

 Filter up to mesh 10 using a median sieve, then up to mesh 100 using an
 opening operator, and output the band pass signal between meshes 11 and 100:

	Y = SIV2d(x, [10 100], 'mo', [0 1], 'b')

 Form a sequence of low pass outputs using closing operators:

	[Y1 Y2 Y3 Y4] = SIV2d(x, [8 16 32 64], 'cccc', [1 2 3 4])

 As above, with initial noise reduction using a median filter:

	[Y1 Y2 Y3 Y4] = SIV2d(x, [4 8 16 32 64], 'mcccc', [0 1 2 3 4])

 Outputting granules (extrema) from an image with a maximum of 64 levels
 and a selection ratio of 4:

	Extrema = SIV2d(x, [4 8 16 32 64], 'mcccc', [0 1 2 3 4], 'g',[64,4])

 FILE
	SIV2d/SIV2d.m

 AUTHOR
	rwy Cambridge Consultants (project leader and algorithms J.A. Bangham, UEA)

 MODIFICATION HISTORY

	1.1	02:jun:95	rwy	First version submitted to SCCS.
	1.2	26:jun:95	rwy	Extended for open, close, M & N sieves.
	1.3	27:jul:95	rwy	Allows bandpass outputs & granule fusing
 	1.4	03:aug:95	rwy	Minor revision.
	1.5	09:aug:95	rwy	Added granule output.
	1.6	14:sep:95	rwy	Added edge map output.
   2.0 01/01/2013  jab Updated to 64 bit and includes additional outputs

 SVN IDENTIFIER
	sieve2d.m	1.6 9/14/95
	SIV2d.m	2.0 01/01/2013


test.m

      [X,Y]=ind2sub(size(imOut),find(imOut>-1));
      X=reshape(X,size(imOut));
      Y=reshape(Y,size(imOut));
      surf(X,Y,double(imOut)/256,'EdgeColor','none');
   Overloaded methods:
      classregtree/test


testCard.m

function [ im, granules, regions ] = testCard( verbose,N,siz, positions,diameters,graded,amplitudeverbose)

N, number of objects
siz, rectangle with edges of length siz
positions, positions as a fraction of siz, N rows of x,y; pairs
diameters, diameters of spots as fraction of siz, N rows. If negative a quarter arc.
amplitude, amplitude as fraction of half intensity (the backround). Minus denonotes minima
graded, true means spots will be intensity cones, false uniform spots

im resulting image
regions, list of N regions
         regions(i).x
         regions(i).y
         regions(i).diameter
Usage
[imOut,granules]=testCard(1,7,[0.5,0.5],0.5,true)

 imOut=5*ones(7,7);imOut(2:3,2)=1;imOut(2,5:6)=10;imOut(4,5:6)=12;imOut(6,4:6)=13;imOut(2,3)=7;imOut(4,2)=7
v=sieve2d(imOut,5,'o',1,'v',4,1);        


verboseSieve2d.m


C:\Users\AB\DArT_Toolshed\Algorithms\Sieve\UpdatedSieve - Copy

displayRegion_Graph_Vertex.m


display_Region_Graph_Vertex.m


extremalTree.m

function [ imOut, granules, regions ] = extremalTree( ( N ,siz, positions,diameters,graded) )

 Display a tree representation of a greyscale im


extremalTreeUtility.m

function [ imOut, granules, regions ] = extremalTreeUtility( ( N ,siz, positions,diameters,graded) )

data.showText

 Display a tree representation of a greyscale im


C:\Users\AB\DArT_Toolshed\Algorithms\Sieve\UpdatedSieve - Copy\sieveCode

testCard.m

function [ im, granules, regions ] = testCard( verbose,N,siz, positions,diameters,graded,amplitudeverbose)

N, number of objects
siz, rectangle with edges of length siz
positions, positions as a fraction of siz, N rows of x,y; pairs
diameters, diameters of spots as fraction of siz, N rows. If negative a quarter arc.
amplitude, amplitude as fraction of half intensity (the backround). Minus denonotes minima
graded, true means spots will be intensity cones, false uniform spots

im resulting image
regions, list of N regions
         regions(i).x
         regions(i).y
         regions(i).diameter
Usage
[imOut,granules]=testCard(1,7,[0.5,0.5],0.5,true)

 imOut=5*ones(7,7);imOut(2:3,2)=1;imOut(2,5:6)=10;imOut(4,5:6)=12;imOut(6,4:6)=13;imOut(2,3)=7;imOut(4,2)=7
v=sieve2d(imOut,5,'o',1,'v',4,1);        


verboseSieve2d.m


C:\Users\AB\DArT_Toolshed\Algorithms\Sieve\UpdatedSieve - Copy (2)

displayRegion_Graph_Vertex.m


display_Region_Graph_Vertex.m


extremalTree.m

function [ imOut, granules, regions ] = extremalTree( ( N ,siz, positions,diameters,graded) )

 Display a tree representation of a greyscale im


extremalTreeUtility.m

function [ imOut, granules, regions ] = extremalTreeUtility( ( N ,siz, positions,diameters,graded) )

data.showText

 Display a tree representation of a greyscale im


C:\Users\AB\DArT_Toolshed\Algorithms\Sieve\UpdatedSieve - Copy (2)\sieveCode

test.m

      [X,Y]=ind2sub(size(imOut),find(imOut>-1));
      X=reshape(X,size(imOut));
      Y=reshape(Y,size(imOut));
      surf(X,Y,double(imOut)/256,'EdgeColor','none');
   Overloaded methods:
      classregtree/test


testCard.m

function [ im, granules, regions ] = testCard( verbose,N,siz, positions,diameters,graded,amplitudeverbose)

N, number of objects
siz, rectangle with edges of length siz
positions, positions as a fraction of siz, N rows of x,y; pairs
diameters, diameters of spots as fraction of siz, N rows. If negative a quarter arc.
amplitude, amplitude as fraction of half intensity (the backround). Minus denonotes minima
graded, true means spots will be intensity cones, false uniform spots

im resulting image
regions, list of N regions
         regions(i).x
         regions(i).y
         regions(i).diameter
Usage
[imOut,granules]=testCard(1,7,[0.5,0.5],0.5,true)

 imOut=5*ones(7,7);imOut(2:3,2)=1;imOut(2,5:6)=10;imOut(4,5:6)=12;imOut(6,4:6)=13;imOut(2,3)=7;imOut(4,2)=7
v=sieve2d(imOut,5,'o',1,'v',4,1);        


verboseSieve2d.m


C:\Users\AB\DArT_Toolshed\Algorithms\Sieve\UpdatedSieve - Copy (3)

displayRegion_Graph_Vertex.m


display_Region_Graph_Vertex.m


extremalTree.m

function [ imOut, granules, regions ] = extremalTree( ( N ,siz, positions,diameters,graded) )

 Display a tree representation of a greyscale im


extremalTreeUtility.m

function [ imOut, granules, regions ] = extremalTreeUtility( ( N ,siz, positions,diameters,graded) )

data.showText

 Display a tree representation of a greyscale im


C:\Users\AB\DArT_Toolshed\Algorithms\Sieve\UpdatedSieve - Copy (3)\sieveCode

test.m

      [X,Y]=ind2sub(size(imOut),find(imOut>-1));
      X=reshape(X,size(imOut));
      Y=reshape(Y,size(imOut));
      surf(X,Y,double(imOut)/256,'EdgeColor','none');
   Overloaded methods:
      classregtree/test


testCard.m

function [ im, granules, regions ] = testCard( verbose,N,siz, positions,diameters,graded,amplitudeverbose)

N, number of objects
siz, rectangle with edges of length siz
positions, positions as a fraction of siz, N rows of x,y; pairs
diameters, diameters of spots as fraction of siz, N rows. If negative a quarter arc.
amplitude, amplitude as fraction of half intensity (the backround). Minus denonotes minima
graded, true means spots will be intensity cones, false uniform spots

im resulting image
regions, list of N regions
         regions(i).x
         regions(i).y
         regions(i).diameter
Usage
[imOut,granules]=testCard(1,7,[0.5,0.5],0.5,true)

 imOut=5*ones(7,7);imOut(2:3,2)=1;imOut(2,5:6)=10;imOut(4,5:6)=12;imOut(6,4:6)=13;imOut(2,3)=7;imOut(4,2)=7
v=sieve2d(imOut,5,'o',1,'v',4,1);        


verboseSieve2d.m


C:\Users\AB\DArT_Toolshed\Algorithms\Sieve\UpdatedSieve - Copy (4)

displayRegion_Graph_Vertex.m


display_Region_Graph_Vertex.m


extremalTree.m

function [ imOut, granules, regions ] = extremalTree( ( N ,siz, positions,diameters,graded) )

 Display a tree representation of a greyscale im


extremalTreeUtility.m

function [ imOut, granules, regions ] = extremalTreeUtility( ( N ,siz, positions,diameters,graded) )

data.showText

 Display a tree representation of a greyscale im


showGranules.m

function showGranules
  
Toggle views of significant granules


C:\Users\AB\DArT_Toolshed\Algorithms\Sieve\UpdatedSieve - Copy (4)\sieveCode

test.m

      [X,Y]=ind2sub(size(imOut),find(imOut>-1));
      X=reshape(X,size(imOut));
      Y=reshape(Y,size(imOut));
      surf(X,Y,double(imOut)/256,'EdgeColor','none');
   Overloaded methods:
      classregtree/test


testCard.m

function [ im, granules, regions ] = testCard( verbose,N,siz, positions,diameters,graded,amplitudeverbose)

N, number of objects
siz, rectangle with edges of length siz
positions, positions as a fraction of siz, N rows of x,y; pairs
diameters, diameters of spots as fraction of siz, N rows. If negative a quarter arc.
amplitude, amplitude as fraction of half intensity (the backround). Minus denonotes minima
graded, true means spots will be intensity cones, false uniform spots

im resulting image
regions, list of N regions
         regions(i).x
         regions(i).y
         regions(i).diameter
Usage
[imOut,granules]=testCard(1,7,[0.5,0.5],0.5,true)

 imOut=5*ones(7,7);imOut(2:3,2)=1;imOut(2,5:6)=10;imOut(4,5:6)=12;imOut(6,4:6)=13;imOut(2,3)=7;imOut(4,2)=7
v=sieve2d(imOut,5,'o',1,'v',4,1);        


verboseSieve2d.m


C:\Users\AB\DArT_Toolshed\Algorithms\Sieve\UpdatedSieve - Copy (5)

ConnectedSets.m


C:\Users\AB\DArT_Toolshed\Algorithms\Sieve\UpdatedSieve - Copy (5)\Images

displayRegion_Graph_Vertex.m


display_Region_Graph_Vertex.m


extremalTree.m

function [ imOut, granules, regions ] = extremalTree( ( N ,siz, positions,diameters,graded) )

 Display a tree representation of a greyscale im


extremalTreeUtility.m

function [ imOut, granules, regions ] = extremalTreeUtility( ( N ,siz, positions,diameters,graded) )

data.showText

 Display a tree representation of a greyscale im


showGranules.m

function showGranules
  
Toggle views of significant granules


C:\Users\AB\DArT_Toolshed\Algorithms\Sieve\UpdatedSieve - Copy (5)\sieveCode

test.m

      [X,Y]=ind2sub(size(imOut),find(imOut>-1));
      X=reshape(X,size(imOut));
      Y=reshape(Y,size(imOut));
      surf(X,Y,double(imOut)/256,'EdgeColor','none');
   Overloaded methods:
      classregtree/test


testCard.m

function [ im, granules, regions ] = testCard( verbose,N,siz, positions,diameters,graded,amplitudeverbose)

N, number of objects
siz, rectangle with edges of length siz
positions, positions as a fraction of siz, N rows of x,y; pairs
diameters, diameters of spots as fraction of siz, N rows. If negative a quarter arc.
amplitude, amplitude as fraction of half intensity (the backround). Minus denonotes minima
graded, true means spots will be intensity cones, false uniform spots

im resulting image
regions, list of N regions
         regions(i).x
         regions(i).y
         regions(i).diameter
Usage
[imOut,granules]=testCard(1,7,[0.5,0.5],0.5,true)

 imOut=5*ones(7,7);imOut(2:3,2)=1;imOut(2,5:6)=10;imOut(4,5:6)=12;imOut(6,4:6)=13;imOut(2,3)=7;imOut(4,2)=7
v=sieve2d(imOut,5,'o',1,'v',4,1);        


verboseSieve2d.m


C:\Users\AB\DArT_Toolshed\Algorithms\Sieve\UpdatedSieve - Copy (6)

ArrowRegions.m


ConnectedSets.m


C:\Users\AB\DArT_Toolshed\Algorithms\Sieve\UpdatedSieve - Copy (6)\Images

C:\Users\AB\DArT_Toolshed\Algorithms\Sieve\UpdatedSieve - Copy (6)\Snapshots

displayRegion_Graph_Vertex.m


display_Region_Graph_Vertex.m


extremalTree.m

function [ imOut, granules, regions ] = extremalTree( ( N ,siz, positions,diameters,graded) )

 Display a tree representation of a greyscale im


extremalTreeUtility.m

function [ imOut, granules, regions ] = extremalTreeUtility( ( N ,siz, positions,diameters,graded) )

data.showText

 Display a tree representation of a greyscale im


showGranules.m

function showGranules
  
Toggle views of significant granules


C:\Users\AB\DArT_Toolshed\Algorithms\Sieve\UpdatedSieve - Copy (6)\sieveCode

test.m

      [X,Y]=ind2sub(size(imOut),find(imOut>-1));
      X=reshape(X,size(imOut));
      Y=reshape(Y,size(imOut));
      surf(X,Y,double(imOut)/256,'EdgeColor','none');
   Overloaded methods:
      classregtree/test


testCard.m

function [ im, granules, regions ] = testCard( verbose,N,siz, positions,diameters,graded,amplitudeverbose)

N, number of objects
siz, rectangle with edges of length siz
positions, positions as a fraction of siz, N rows of x,y; pairs
diameters, diameters of spots as fraction of siz, N rows. If negative a quarter arc.
amplitude, amplitude as fraction of half intensity (the backround). Minus denonotes minima
graded, true means spots will be intensity cones, false uniform spots

im resulting image
regions, list of N regions
         regions(i).x
         regions(i).y
         regions(i).diameter
Usage
[imOut,granules]=testCard(1,7,[0.5,0.5],0.5,true)

 imOut=5*ones(7,7);imOut(2:3,2)=1;imOut(2,5:6)=10;imOut(4,5:6)=12;imOut(6,4:6)=13;imOut(2,3)=7;imOut(4,2)=7
v=sieve2d(imOut,5,'o',1,'v',4,1);        


verboseSieve2d.m


C:\Users\AB\DArT_Toolshed\Algorithms\Sieve\UpdatedSieve - Copy (7)

ArrowRegions.m


ConnectedSets.m


C:\Users\AB\DArT_Toolshed\Algorithms\Sieve\UpdatedSieve - Copy (7)\Images

C:\Users\AB\DArT_Toolshed\Algorithms\Sieve\UpdatedSieve - Copy (7)\Snapshots

displayRegion_Graph_Vertex.m


display_Region_Graph_Vertex.m


extremalTree.m

function [ imOut, granules, regions ] = extremalTree( ( N ,siz, positions,diameters,graded) )

 Display a tree representation of a greyscale im


extremalTreeFig.m


extremalTreeUtility.m

function [ imOut, granules, regions ] = extremalTreeUtility( ( N ,siz, positions,diameters,graded) )

data.showText

 Display a tree representation of a greyscale im


extremalTreeUtility2.m

function [ imOut, granules, regions ] = extremalTreeUtility2( ( N ,siz, positions,diameters,graded) )

 To be compatible with ConnectedSets.m (i.e. the plotting axis)

data.showText

 Display a tree representation of a greyscale im


showGranules.m

function showGranules
  
Toggle views of significant granules


C:\Users\AB\DArT_Toolshed\Algorithms\Sieve\UpdatedSieve - Copy (7)\sieveCode

test.m

      [X,Y]=ind2sub(size(imOut),find(imOut>-1));
      X=reshape(X,size(imOut));
      Y=reshape(Y,size(imOut));
      surf(X,Y,double(imOut)/256,'EdgeColor','none');
   Overloaded methods:
      classregtree/test


testCard.m

function [ im, granules, regions ] = testCard( verbose,N,siz, positions,diameters,graded,amplitudeverbose)

N, number of objects
siz, rectangle with edges of length siz
positions, positions as a fraction of siz, N rows of x,y; pairs
diameters, diameters of spots as fraction of siz, N rows. If negative a quarter arc.
amplitude, amplitude as fraction of half intensity (the backround). Minus denonotes minima
graded, true means spots will be intensity cones, false uniform spots

im resulting image
regions, list of N regions
         regions(i).x
         regions(i).y
         regions(i).diameter
Usage
[imOut,granules]=testCard(1,7,[0.5,0.5],0.5,true)

 imOut=5*ones(7,7);imOut(2:3,2)=1;imOut(2,5:6)=10;imOut(4,5:6)=12;imOut(6,4:6)=13;imOut(2,3)=7;imOut(4,2)=7
v=sieve2d(imOut,5,'o',1,'v',4,1);        


verboseSieve2d.m


C:\Users\AB\DArT_Toolshed\Algorithms\Sieve\UpdatedSieve - Copy (8)

ArrowRegions.m


ConnectedSets.m


MakeQuantiserTable.m


QuantiseDensity.m


C:\Users\AB\DArT_Toolshed\Algorithms\Sieve\UpdatedSieve - Copy (8)\SIV_TestC_20130410

C:\Users\AB\DArT_Toolshed\Algorithms\Sieve\UpdatedSieve - Copy (8)\SIV_TestC_20130410\directory_of_images_for_SIV

C:\Users\AB\DArT_Toolshed\Algorithms\Sieve\UpdatedSieve - Copy (8)\SIV_TestD_20130410

C:\Users\AB\DArT_Toolshed\Algorithms\Sieve\UpdatedSieve - Copy (8)\SIV_TestD_20130410\directory_of_images_for_SIV

C:\Users\AB\DArT_Toolshed\Algorithms\Sieve\UpdatedSieve - Copy (8)\SIV_testA_20130410

C:\Users\AB\DArT_Toolshed\Algorithms\Sieve\UpdatedSieve - Copy (8)\SIV_testB_20130410

C:\Users\AB\DArT_Toolshed\Algorithms\Sieve\UpdatedSieve - Copy (8)\SIV_testB_20130410\directory_of_images_for_SIV

C:\Users\AB\DArT_Toolshed\Algorithms\Sieve\UpdatedSieve - Copy (8)\SIV_test_20130410

SieveCTScans.m

 Program name: SieveCTScans
 Authors: GDT + PS
 Date:     15/09/2011


C:\Users\AB\DArT_Toolshed\Algorithms\Sieve\UpdatedSieve - Copy (8)\Snapshots

checkbox_isometric.m


C:\Users\AB\DArT_Toolshed\Algorithms\Sieve\UpdatedSieve - Copy (8)\directory_of_images_for_SIV

displayRegion_Graph_Vertex.m


display_Region_Graph_Vertex.m


extremalMSERUtility2.m

function [ imOut, granules, regions ] = extremalMSERUtility2( ( N ,siz, positions,diameters,graded) )

 To be compatible with ConnectedSets.m (i.e. the plotting axis)

data.showText

 Display a tree representation of a greyscale im


extremalTree.m

function [ imOut, granules, regions ] = extremalTree( ( N ,siz, positions,diameters,graded) )

 Display a tree representation of a greyscale im


extremalTreeFig.m


extremalTreeUtility.m

function [ imOut, granules, regions ] = extremalTreeUtility( ( N ,siz, positions,diameters,graded) )

data.showText

 Display a tree representation of a greyscale im


extremalTreeUtility2 - Copy.m

function [ imOut, granules, regions ] = extremalTreeUtility2( ( N ,siz, positions,diameters,graded) )

 To be compatible with ConnectedSets.m (i.e. the plotting axis)

data.showText

 Display a tree representation of a greyscale im


extremalTreeUtility2.m

function [ imOut, granules, regions ] = extremalTreeUtility2( ( N ,siz, positions,diameters,graded) )

 To be compatible with ConnectedSets.m (i.e. the plotting axis)

data.showText

 Display a tree representation of a greyscale im


quantise_uint8_im.m


saveCurrentStateSIV.m


showGranules.m

function showGranules
  
Toggle views of significant granules


showTheText.m


C:\Users\AB\DArT_Toolshed\Algorithms\Sieve\UpdatedSieve - Copy (8)\sieveCode

sieveSect.m

 Program name: SIEVESECT
 Authors: GDT + PS
 Date:     15/09/2011


siv2d_m.m

 [Y1 Y2 ... Ym] = siv2d_m(X, [M1 M2 ... Mn], 'F1F2 ... Fn',
						[O1 O2 ... On], type, N, Levels)

 CALLS the mex code SIV2d.* To get a sequence of outputs call SIVE2d directly

 Perform a 2-D connected-set datasieve based on area.

 X	    :	uint8, input image
 Mk	:	maximum mesh (scale, area of extrema) for each filter
 Fk	:	filter type  (m, o, c, M or N)	[default: m]
 Ok	: 	output index for each filter    [default: 1 for last filter]
 type  :	type of output  (l, b, e, g, f or v)	[default: l]
 N	    :	connectivity (4 or 8)           [default: 4]
 EP    :   parameters for selecting extrema [Levels,Ratio]
              Levels :   number of finite levels in image [default: 256]
              Ratio  :   typically 5 [default: 1]
              MinArea:   typically 4 [default: 1]
              MaxArea:   typically 1000 [default: 1000]
              MinAmp :   typically 4 [default: 0]

 Yj	:	output image(j) from a filter

 FILTERS
 The input image is processed using one or more filters Fk, applied
 sequentially, up to the associated maximum meshes Mk.  The available
 filters are:

 m     :	median     (maxima & minima processed in arbitrary order)
 o     : 	opening    (only maxima processed, c.f. maximally stable extreme regions )
 c     : 	closing    (only minima processed, c.f. maximally stable extreme regions)
 M     : 	open-close (maxima then minima processed at each scale)
 N     : 	close-open (minima then maxima processed at each scale)

 For each filter, the result may optionally be output by specifying the
 index of the desired output argument as the input Ok (where Ok=1 represents
 the first output argument).  If Ok=0 then no output is generated
 for filter Fk.

 TYPE OF OUTPUT
 The available output types are:

 l	:	low-pass
 b	:	band-pass
 f	:	band-pass with granule fusing
 e :	band-pass, returning edge map
 g	:	granule list
 v :   verbose, i.e. main data structure of engine

 where the band-pass output is the difference between the outputs of the
 current filter and the previous filter (or the input image) in the
 sequence of filters.  The same output type must be used for each filter.

 The granule output list format is a cell array containing the following data:
 For example:
            Number: 3 (number of granules)
              area: [1 5 9]
             value: [18 7 11]
             level: [255 237 230]
         deltaArea: [4 4 4]
         last_area: [400 80 44]
              root: [76 76 76]
    PictureElement: {[77]  [77 94 60 78 76]  [77 95 61 93 59 94 60 78 76]}

 e.g.  data.g.PictureElement{3} contains indexes to the pixels that formed the granule

 Note that producing multiple low or band pass outputs in this way is much
 more efficient than calling the sieve function several times with different
 meshes.

 Memory requirements are about 18 bytes per image pel.


 EXAMPLES:

 Filter up to mesh 100 using a median sieve, and generate a low pass output:

 	Y = SIV2d(x, 100)
 or 	Y = SIV2d(x, 100, 'm')
 or	Y = SIV2d(x, 100, 'm', 1)
 or	Y = SIV2d(x, 100, 'm', 1, 'l')
 or	Y = SIV2d(x, 100, 'm', 1, 'l',4)
 or	Y = SIV2d(x, 100, 'm', 1, 'l',4,[256,4,5])

 Filter up to mesh 10 using a median sieve, then up to mesh 100 using an
 opening operator, and output the band pass signal between meshes 11 and 100:

	Y = SIV2d(x, [10 100], 'mo', [0 1], 'b')

 Form a sequence of low pass outputs using closing operators:

	[Y1 Y2 Y3 Y4] = SIV2d(x, [8 16 32 64], 'cccc', [1 2 3 4])

 As above, with initial noise reduction using a median filter:

	[Y1 Y2 Y3 Y4] = SIV2d(x, [4 8 16 32 64], 'mcccc', [0 1 2 3 4])

 Outputting granules (extrema) from an image with a maximum of 64 levels
 and a selection ratio of 4:

	Extrema = SIV2d(x, [4 8 16 32 64], 'mcccc', [0 1 2 3 4], 'g',[64,4])

 FILE
	SIV2d/SIV2d.m

 AUTHOR
	rwy Cambridge Consultants (project leader and algorithms J.A. Bangham, UEA)

 MODIFICATION HISTORY

	1.1	02:jun:95	rwy	First version submitted to SCCS.
	1.2	26:jun:95	rwy	Extended for open, close, M & N sieves.
	1.3	27:jul:95	rwy	Allows bandpass outputs & granule fusing
 	1.4	03:aug:95	rwy	Minor revision.
	1.5	09:aug:95	rwy	Added granule output.
	1.6	14:sep:95	rwy	Added edge map output.
   2.0 01/01/2013  jab Updated to 64 bit and includes additional outputs

 SVN IDENTIFIER
	sieve2d.m	1.6 9/14/95
	SIV2d.m	2.0 01/01/2013


test.m

      [X,Y]=ind2sub(size(imOut),find(imOut>-1));
      X=reshape(X,size(imOut));
      Y=reshape(Y,size(imOut));
      surf(X,Y,double(imOut)/256,'EdgeColor','none');
   Overloaded methods:
      classregtree/test


testCard.m

function [ im, granules, regions ] = testCard( verbose,N,siz, positions,diameters,graded,amplitudeverbose)

N, number of objects
siz, rectangle with edges of length siz
positions, positions as a fraction of siz, N rows of x,y; pairs
diameters, diameters of spots as fraction of siz, N rows. If negative a quarter arc.
amplitude, amplitude as fraction of half intensity (the backround). Minus denonotes minima
graded, true means spots will be intensity cones, false uniform spots

im resulting image
regions, list of N regions
         regions(i).x
         regions(i).y
         regions(i).diameter
Usage
[imOut,granules]=testCard(1,7,[0.5,0.5],0.5,true)

 imOut=5*ones(7,7);imOut(2:3,2)=1;imOut(2,5:6)=10;imOut(4,5:6)=12;imOut(6,4:6)=13;imOut(2,3)=7;imOut(4,2)=7
v=sieve2d(imOut,5,'o',1,'v',4,1);        


verboseSieve2d.m