Home > utils > uimage.m

uimage

PURPOSE ^

UIMAGE Display image with uneven axis.

SYNOPSIS ^

function h = uimage(varargin)

DESCRIPTION ^

UIMAGE  Display image with uneven axis.
   UIMAGE(X,Y,C) displays matrix C as an image, using the vectors X and
   Y to specify the X and Y coordinates. X and Y may be unevenly spaced
   vectors, but must be increasing. The size of C must be LENGTH(Y)*
   LENGTH(X). (Most probably you'll want to display C' instead of C).

   Contrary to Matlab's original IMAGE function, here the vectors X and Y
   do not need to be linearly spaced. Whereas IMAGE linearly interpolates
   the X-axis between X(1) and X(end), ignoring all other values (idem
   for Y), UIMAGE allows for X and/or Y to be unevenly spaced vectors, by
   locally stretching the matrix C (ie, by duplicating some elements of C)
   for larger X and/or Y intervals.

   The syntax for UIMAGE(X,Y,C,...) is the same as IMAGE(X,Y,C,...)
   (all the remaining arguments, eg 'PropertyName'-PropertyValue pairs,
   are passed to IMAGE). See IMAGE for details.

   Use UIMAGESC to scale the data using the full colormap. The syntax for
   UIMAGESC(X,Y,C,...) is the same as IMAGESC(X,Y,C,...).

   Typical uses:
      - Plotting a spatio-temporal diagram (T,X), with unevenly spaced
      time intervals for T (eg, when some values are missing, or when
      using a non-constant sampling rate).
      - Plotting a set of power spectra with frequency in log-scale.

   h = UIMAGE(X,Y,C,...) returns a handle to the image.

   Example:
     c = randn(50,20);         % Random 50x20 matrix
     x = logspace(1,3,50);     % log-spaced X-axis, between 10 and 1000
     y = linspace(3,8,20);     % lin-spaced Y-axis, between 3 and 8
     uimagesc(x,y,c');         % displays the matrix

   F. Moisy
   Revision: 1.03,  Date: 2006/06/14.

   See also IMAGE, IMAGESC, UIMAGESC.

CROSS-REFERENCE INFORMATION ^

This function calls: This function is called by:

SOURCE CODE ^

0001 function h = uimage(varargin)
0002 %UIMAGE  Display image with uneven axis.
0003 %   UIMAGE(X,Y,C) displays matrix C as an image, using the vectors X and
0004 %   Y to specify the X and Y coordinates. X and Y may be unevenly spaced
0005 %   vectors, but must be increasing. The size of C must be LENGTH(Y)*
0006 %   LENGTH(X). (Most probably you'll want to display C' instead of C).
0007 %
0008 %   Contrary to Matlab's original IMAGE function, here the vectors X and Y
0009 %   do not need to be linearly spaced. Whereas IMAGE linearly interpolates
0010 %   the X-axis between X(1) and X(end), ignoring all other values (idem
0011 %   for Y), UIMAGE allows for X and/or Y to be unevenly spaced vectors, by
0012 %   locally stretching the matrix C (ie, by duplicating some elements of C)
0013 %   for larger X and/or Y intervals.
0014 %
0015 %   The syntax for UIMAGE(X,Y,C,...) is the same as IMAGE(X,Y,C,...)
0016 %   (all the remaining arguments, eg 'PropertyName'-PropertyValue pairs,
0017 %   are passed to IMAGE). See IMAGE for details.
0018 %
0019 %   Use UIMAGESC to scale the data using the full colormap. The syntax for
0020 %   UIMAGESC(X,Y,C,...) is the same as IMAGESC(X,Y,C,...).
0021 %
0022 %   Typical uses:
0023 %      - Plotting a spatio-temporal diagram (T,X), with unevenly spaced
0024 %      time intervals for T (eg, when some values are missing, or when
0025 %      using a non-constant sampling rate).
0026 %      - Plotting a set of power spectra with frequency in log-scale.
0027 %
0028 %   h = UIMAGE(X,Y,C,...) returns a handle to the image.
0029 %
0030 %   Example:
0031 %     c = randn(50,20);         % Random 50x20 matrix
0032 %     x = logspace(1,3,50);     % log-spaced X-axis, between 10 and 1000
0033 %     y = linspace(3,8,20);     % lin-spaced Y-axis, between 3 and 8
0034 %     uimagesc(x,y,c');         % displays the matrix
0035 %
0036 %   F. Moisy
0037 %   Revision: 1.03,  Date: 2006/06/14.
0038 %
0039 %   See also IMAGE, IMAGESC, UIMAGESC.
0040 
0041 
0042 % History:
0043 % 2006/06/12: v1.00, first version.
0044 % 2006/06/14: v1.03, minor bug fixed; works in ML6.
0045 % 2006/12/14 - Petr Janata modified algorithm for determining if dimension is evenly spaced
0046 
0047 error(nargchk(3,inf,nargin));
0048 
0049 % maximum number of matrix elements to interpolate the uneven axis
0050 % (typically between 500 and 5000):
0051 nmax = 2000; 
0052 
0053 x = varargin{1};
0054 y = varargin{2};
0055 c = varargin{3};
0056 
0057 if any(diff(x)<=0) || any(diff(y)<=0)
0058     error('The X and Y axis should be increasing.');
0059 end
0060 
0061 dx = min(diff(x));                   % smallest interval for X
0062 dy = min(diff(y));                   % smallest interval for Y
0063 
0064 % test if X and Y are linearly spaced (to within 10^-12):
0065 evenx = all(abs(diff(x)/dx-1)<1e-12);     % true if X is linearly spaced
0066 evenx = all(abs(diff(diff(x)))<1e-12);    % PJ
0067 eveny = all(abs(diff(y)/dy-1)<1e-12);     % true if Y is linearly spaced
0068 eveny = all(abs(diff(diff(y)))<1e-12);    % PJ
0069 
0070 
0071 if evenx && eveny         % X and Y both evenly spaced
0072 
0073     xe = x;
0074     ye = y;
0075     ce = c;
0076 
0077 elseif evenx && ~eveny    % X even and Y uneven
0078     
0079     nx = length(x);
0080     xe = x;
0081 
0082     ny = ceil(1 + (y(end) - y(1))/dy);   % number of points for Y
0083     ny = min(ny, nmax);
0084     ye = linspace(y(1), y(end), ny);
0085 
0086     ce = zeros(ny,nx);
0087 
0088     for j=1:ny
0089         indj = find(y<=ye(j));
0090         ce(j,1:nx) = c(indj(end), 1:nx);
0091     end;
0092 
0093 elseif ~evenx && eveny    % X uneven and Y even
0094     
0095     nx = ceil(1 + (x(end) - x(1))/dx);   % number of points for X
0096     nx = min(nx, nmax);
0097     xe = linspace(x(1), x(end), nx);
0098 
0099     ny = length(y);
0100     ye = y;
0101 
0102     ce = zeros(ny,nx);
0103 
0104     for i=1:nx
0105         indi = find(x<=xe(i));
0106         ce(1:ny,i) = c(1:ny, indi(end));
0107     end;
0108 
0109 elseif ~evenx && ~eveny   % X and Y both uneven
0110     
0111     nx = ceil(1 + (x(end) - x(1))/dx);   % number of points for X
0112     nx = min(nx, nmax);
0113     xe = linspace(x(1), x(end), nx);
0114 
0115     ny = ceil(1 + (y(end) - y(1))/dy);   % number of points for Y
0116     ny = min(ny, nmax);
0117     ye = linspace(y(1), y(end), ny);
0118 
0119     ce = zeros(ny,nx);
0120 
0121     for i=1:nx
0122         for j=1:ny
0123             indi = find(x<=xe(i));
0124             indj = find(y<=ye(j));
0125             ce(j,i) = c(indi(end), indj(end));
0126         end;
0127     end;
0128 
0129 end
0130 
0131 hh = image(xe, ye, ce, varargin{4:end});
0132 
0133 if nargout>0
0134     h = hh;
0135 end

Generated on Tue 16-Jul-2019 04:00:51 by m2html © 2003