How to plot a graph of a custom surface in Matlab? - matlab

I would like to plot the surface of this function in Matlab:
w=(m*(1-p))/(h-1+m*(1-p))
Where
h=0.98;
P=[0.05:0.001:0.98];
M=[0:0.001:1];
The graph should look like this:
I tried to implement the graphing function as:
h=0.98;
P=[0.05:0.001:0.98];
M=[0:0.001:1];
W=[];
for i=1:size(P, 2)
for j=1:size(M, 2)
p=P(i);
m=M(j);
w=(m*(1-p))/(h-1+m*(1-p));
if w>1000
w=1000;
end
if w<0
w=0;
end
W(i, j) = w;
end
end
mesh(M, P, W);
But since the function changes rapidly around the boundary condition, my graph ended up looking like this:
How can I smooth out the curve and give it a nice, uniform grid like in the first picture?

Here's a way to do what you want. See comments in the code for more information.
%% //Definitions
w = #(h,m,p)(m.*(1-p))./(h-1+m.*(1-p)); %// w as a function handle
w_min = 0; w_max = 10; %// For plots
cMap = hot(200); cMap = cMap(floor(0.5*length(cMap)):ceil(0.7*length(cMap)),:);
h = 0.98;
m_start = 0; m_end = 1; m_step = [0.02 0.01 0.001];
p_start = 0.05; p_end = 0.98; p_step = [0.1 0.01 0.001];
%// Definitions of the mesh used in plotting (same for all charts)
mesh_m_step = 0.05; mesh_p_step = 0.05;
[meshM,meshP] = meshgrid(m_start:mesh_m_step:m_end,p_start:mesh_p_step:p_end);
meshW = w(h,meshM,meshP);
%% //Evaluation and plotting:
figure(); set(gcf,'Position',[240,500,1391,376]); subplot(1,3,1);
for ind1 = 1:3
%% // Evaluate equation
[MM,PP] = meshgrid(m_start:m_step(ind1):m_end,p_start:p_step(ind1):p_end);
WW = w(h,MM,PP);
%% // Plot
subplot(1,3,ind1);
surf(MM,PP,WW,'EdgeColor','none'); view([45,45]);
hold on; mesh(meshM,meshP,meshW,'EdgeColor','k','FaceColor','none');
colormap(cMap); %// Fix the colormap
zlim([w_min,w_max]); caxis(zlim) %// Modify zlim
end
Which results in:

Related

coloring 3D plots on MATLAB

I made two 3D plots on the same axis. now I desire to give them different colors for easy identification. How do I do this coloring? The MATLAB code is shown below.
tic
Nx = 50;
Ny = 50;
x = linspace(0,1,Nx);
y = linspace(0,0.5,Ny);
[X,Y] = meshgrid(x,y);
[M,N] = size(X);
for m=1:M
for n=1:N
%get x,y coordinate
x_mn = X(m,n);
y_mn = Y(m,n);
%%% X=D2 and Y=D1
%Check if x_mn and y_mn satisfy requirement
if(x_mn >= y_mn)
%evaluate function 1
Z(m,n) = (x_mn^2 - 2*x_mn*y_mn + y_mn^2);
Z_1(m,n) = (x_mn^2);
elseif(x_mn < y_mn)
%evaluate function 2
Z(m,n) = 0;
Z_1(m,n) = (x_mn^2);
%% Z(m,n) = 2*(2*x_mn*y_mn + y_mn - y_mn^2 - 2*x_mn);
else
Z(m,n) = 0;
end
end
end
%Plot the surface
figure
surf(X,Y,Z) %first plot
surfc(X,Y,Z)
hold on
surf(X,Y,Z_1) %second plot
xlabel('Dm');
ylabel('D');
zlabel('pR');
grid on
shading interp
toc
disp('DONE!')
How can I create two differently colored surfaces?
figure
surf(X,Y,Z) %first plot
surfc(X,Y,Z)
hold on
surf(X,Y,Z_1)
Your surfc() call actually overwrites your surf() call, is this intended?
As to your colour: the documentation is a marvellous thing:
surfc(X,Y,Z,C) additionally specifies the surface color.
In other words: just specify the colour as you want it. C needs to be a matrix of size(Z) with the desired colours, i.e. set all of them equal to create an monocoloured surface:
x = 1:100;
y = 1:100;
z = rand(100);
figure;
surfc(x,y,z,ones(size(z)))
hold on
surfc(x,y,z+6,ones(size(z))+4)
Results in (MATLAB R2007b, but the syntax is the same nowadays)

gif of sphere with changing radius in Matlab

I wish to generate a .gif file of a sphere that grows in size(radius) over time based on a specified function relating radius and time. I'm having some trouble with how to formulate the animation.
Here is what I have so far:
%% Parameters
dt = 0.05;
time = 0:dt:1;
radius = 1
%% generate sphere
[X, Y, Z] = sphere(25);
X=X*radius;
Y=Y*radius;
Z=Z*radius;
mySphere = surf(X,Y,Z, 'FaceLighting','gouraud');
axis equal
shading interp
mySphere.FaceAlpha = 0.3
view([61 15])
colormap bone
hold on
%% generate gif
filename = 'Sizechange.gif';
for n = 1:20
radius = 1 + time(n)
im = frame2im(getframe(1));
[imind,cm] = rgb2ind(im,256);
if n == 1;
imwrite(imind,cm,filename,'gif', 'Loopcount',inf,'DelayTime',dt);
else
imwrite(imind,cm,filename,'gif','WriteMode','append','DelayTime',dt);
end
end
Here I am trying to get it to go from radius 1 to radius 2 in steps of 0.05.
When I run this however, the gif stays still at 1 and there is no animation.
Any help would be greatly appreciated.
As #cris-luengo said, you should redraw your sphere for each iteration on the radius.
%% Parameters
dt = 0.05;
time = 0:dt:1;
radius = 1 ;
%% generate sphere
[X, Y, Z] = sphere(25);
X=X*radius;
Y=Y*radius;
Z=Z*radius;
%figure;
%mySphere = surf(X,Y,Z, 'FaceLighting','gouraud');
% axis equal
% shading interp
% mySphere.FaceAlpha = 0.3;
% view([61 15])
% colormap bone
% hold on
%% generate gif
filename = 'Sizechange.gif';
figure;
for n = 1:20
radius = 1+ time(n);
%====================================================
X=X*radius;
Y=Y*radius;
Z=Z*radius;
mySphere = surf(X,Y,Z, 'FaceLighting','gouraud');
axis equal
shading interp
mySphere.FaceAlpha = 0.3;
view([61 15])
colormap bone
%====================================================
im = frame2im(getframe(1));
[imind,cm] = rgb2ind(im,256);
if n == 1
imwrite(imind,cm,filename,'gif', 'Loopcount',inf,'DelayTime',dt);
else
imwrite(imind,cm,filename,'gif','WriteMode','append','DelayTime',dt);
end
end

How can I make a cylindrical 3D contour plot in Matlab?

I have an axisymmetric flow with m x n grid points in r and z direction and I want to plot the temperature that is stored in a matrix with size mxn in a 3D cylindrical plot as visualized in the link below (My reputation is not high enough to include it as a picture).
I have managed to plot it in 2D (r,z plane) using contour but I would like to add the theta plane for visualization. How can I do this?
You can roll your own with multiple calls to surface().
Key idea is: for each surface: (1) theta=theta1, (2) theta=theta2, (3) z=zmax, (4) z=0, (5) r=rmax, generate a 3D mesh (xx,yy,zz) and the temperature map on that mesh. So you have to think about how to construct each surface mesh.
Edit: completed code is now provided. All magic number and fake data are put at (almost) the top of the code so it's easy to convert it into a general purpose Matlab function. Good luck!
% I have adjusted the range values to show the curved cylinder wall
% display a variable temperature
r = 0:0.1:2.6; % you can also try r = 0:0.1:3.0
z = 0:0.1:10; % you can also try z = 0:0.1:15;
[rr, zz] = meshgrid(r,z);
% fake temperature data
temp = 100 + (10* (3-rr).^0.6) .* (1-((zz - 7.5)/7.5).^6) ;
% visualize in 2D
figure(1);
clf;
imagesc(r,z,temp);
colorbar;
% set cut planes angles
theta1 = 0;
theta2 = pi*135/180;
nt = 40; % angle resolution
figure(2);
clf;
xx1 = rr * cos(theta1);
yy1 = rr * sin(theta1);
h1 = surface(xx1,yy1,zz,temp,'EdgeColor', 'none');
xx2 = rr * cos(theta2);
yy2 = rr * sin(theta2);
h2 = surface(xx2,yy2,zz,temp,'EdgeColor', 'none');
% polar meshgrid for the top end-cap
t3 = linspace(theta1, (theta2 - 2*pi), nt);
[rr3, tt3] = meshgrid(r,t3);
xx3 = rr3 .* cos(tt3);
yy3 = rr3 .* sin(tt3);
zz3 = ones(size(rr3)) * max(z);
temp3 = zeros(size(rr3));
for k = 1:length(r)
temp3(:,k) = temp(end,k);
end
h3 = surface(xx3,yy3,zz3,temp3,'EdgeColor', 'none');
% polar meshgrid for the bottom end-cap
zz4 = ones(size(rr3)) * min(z);
temp4 = zeros(size(rr3));
for k = 1:length(r)
temp4(:,k) = temp(1,k);
end
h4 = surface(xx3,yy3,zz4,temp4,'EdgeColor', 'none');
% generate a curved meshgrid
[tt5, zz5] = meshgrid(t3,z);
xx5 = r(end) * cos(tt5);
yy5 = r(end) * sin(tt5);
temp5 = zeros(size(xx5));
for k = 1:length(z)
temp5(k,:) = temp(k,end);
end
h5 = surface(xx5, yy5, zz5,temp5,'EdgeColor', 'none');
axis equal
colorbar
view(125,25); % viewing angles

Ordered Isoline Calculation from 3D Triangular Surface in MATLAB

I need to extract the isoline coordinates of a 4D variable from a 3D surface defined using a triangulated mesh in MATLAB. I need the isoline coordinates to be a ordered in such a manner that if they were followed in order they would trace the path i.e. the order of the points a 3D printer would follow.
I have found a function that can calculate the coordinates of these isolines (see Isoline function here) but the problem is this function does not consider the isolines to be joined in the correct order and is instead a series of 2 points separated by a Nan value. This makes this function only suitable for visualisation purposes and not the path to follow.
Here is a MWE of the problem of a simplified problem, the surface I'm applying it too is much more complex and I cannot share it. Where x, y and z are nodes, with TRI providing the element connectivity list and v is the variable of which I want the isolines extracted from and is not equal to z.
If anyone has any idea on either.....
A function to extract isoline values in the correct order for a 3D tri mesh.
How to sort the data given by the function Isoline so that they are in the correct order.
.... it would be very much appreciated.
Here is the MWE,
% Create coordinates
[x y] = meshgrid( -10:0.5:10, -10:0.5:10 );
z = (x.^2 + y.^2)/20; % Z height
v = x+y; % 4th dimension value
% Reshape coordinates into list to be converted to tri mesh
x = reshape(x,[],1); y = reshape(y,[],1); z = reshape(z,[],1); v = reshape(v,[],1);
TRI = delaunay(x,y); % Convertion to a tri mesh
% This function calculates the isoline coordinates
[xTows, yTows, zTows] = IsoLine( {TRI,[x, y, z]}, v, -18:2:18);
% Plotting
figure(1); clf(1)
subplot(1,2,1)
trisurf(TRI,x,y,z,v)
hold on
for i = 1:size(xTows,1)
plot3( xTows{i,1}, yTows{i,1}, zTows{i,1}, '-k')
end
hold off
shading interp
xlabel('x'); ylabel('y'); zlabel('z'); title('Isolines'), axis equal
%% This section is solely to show that the isolines are not in order
for i = 1:size(xTows,1)
% Arranging data into colums and getting rid of Nans that appear
xb = xTows{i,1}; yb = yTows{i,1}; zb = zTows{i,1};
xb = reshape(xb, 3, [])'; xb(:,3) = [];
yb = reshape(yb, 3, [])'; yb(:,3) = [];
zb = reshape(zb, 3, [])'; zb(:,3) = [];
subplot(1,2,2)
trisurf(TRI,x,y,z,v)
shading interp
view(2)
xlabel('x'); ylabel('y'); zlabel('z'); title('Plotting Isolines in Order')
axis equal; axis tight; hold on
for i = 1:size(xb,1)
plot3( [xb(i,1) xb(i,2)], [yb(i,1) yb(i,2)], [zb(i,1) zb(i,2)], '-k')
drawnow
end
end
and here is the function Isoline, which I have slightly adpated.
function [xTows, yTows, zTows] = IsoLine(Surf,F,V,Col)
if length(Surf)==3 % convert mesh to triangulation
P = [Surf{1}(:) Surf{2}(:) Surf{3}(:)];
Surf{1}(end,:) = 1i;
Surf{1}(:,end) = 1i;
i = find(~imag(Surf{1}(:)));
n = size(Surf{1},1);
T = [i i+1 i+n; i+1 i+n+1 i+n];
else
T = Surf{1};
P = Surf{2};
end
f = F(T(:));
if nargin==2
V = linspace(min(f),max(f),22);
V = V(2:end-1);
elseif numel(V)==1
V = linspace(min(f),max(f),V+2);
V = V(2:end-1);
end
if nargin<4
Col = 'k';
end
H = NaN + V(:);
q = [1:3 1:3];
% -------------------------------------------------------------------------
% Loop over iso-values ----------------------------------------------------
xTows = [];
yTows = [];
zTows = [];
for k = 1:numel(V)
R = {[],[]};
G = F(T) - V(k);
C = 1./(1-G./G(:,[2 3 1]));
f = unique(T(~isfinite(C))); % remove degeneracies by random perturbation
F(f) = F(f).*(1+1e-12*rand(size(F(f)))) + 1e-12*rand(size(F(f)));
G = F(T) - V(k);
C = 1./(1-G./G(:,[2 3 1]));
C(C<0|C>1) = -1;
% process active triangles
for i = 1:3
f = any(C>=0,2) & C(:,i)<0;
for j = i+1:i+2
w = C(f,q([j j j]));
R{j-i} = [R{j-i}; w.*P(T(f,q(j)),:)+(1-w).*P(T(f,q(j+1)),:)];
end
end
% define isoline
for i = 1:3
X{i} = [R{1}(:,i) R{2}(:,i) nan+R{1}(:,i)]';
% X{i} = [R{1}(:,i) R{2}(:,i)]'; % Changed by Matt
X{i} = X{i}(:)';
end
% plot isoline
if ~isempty(R{1})
% hold on
% H(k) = plot3(X{1},X{2},X{3},Col);
% Added by M.Thomas
xTows{k,1} = X{1};
yTows{k,1} = X{2};
zTows{k,1} = X{3};
end
end
What you will notice is that the isolines (xTows, yTows and zTows) are not in order there "jump around" when plotted sequentially. I need to sort the tows so that they give a smooth plot in order.

hough transform for lines

I'm trying to get a Hough transform to work in MATLAB, but I'm having problems. I have a really bad way of detecting peaks that needs to be fixed, but before that I need to be able to reverse the hough transform to create the lines again properly. This is the type of stuff I'm getting right now:
looks like its rotated by 90 degrees, but I'm not sure why. I'm not sure if it's my hough space that's wrong, or if it's the way I dehough and draw the lines. Also could someone help improve my peak detection?
the images used in the code are here
Thank you
%% load a sample image; convert to grayscale; convert to binary
%create 'x' image (works well)
a = eye(255);
b = flipud(eye(255));
x = a + b;
x(128,128) = 1;
%image = rgb2gray(imread('up.png')) < 255;
%image = rgb2gray(imread('hexagon.png')) < 255;
%image = rgb2gray(imread('traingle.png')) < 255;
%%% these work
%image = x;
%image = a;
image = b;
%% set up variables for hough transform
theta_sample_frequency = 0.01;
[x, y] = size(image);
rho_limit = norm([x y]);
rho = (-rho_limit:1:rho_limit);
theta = (0:theta_sample_frequency:pi);
num_thetas = numel(theta);
num_rhos = numel(rho);
hough_space = zeros(num_rhos, num_thetas);
%% perform hough transform
for xi = 1:x
for yj = 1:y
if image(xi, yj) == 1
for theta_index = 1:num_thetas
th = theta(theta_index);
r = xi * cos(th) + yj * sin(th);
rho_index = round(r + num_rhos/2);
hough_space(rho_index, theta_index) = ...
hough_space(rho_index, theta_index) + 1;
end
end
end
end
%% show hough transform
subplot(1,2,1);
imagesc(theta, rho, hough_space);
title('Hough Transform');
xlabel('Theta (radians)');
ylabel('Rho (pixels)');
colormap('gray');
%% detect peaks in hough transform
r = [];
c = [];
[max_in_col, row_number] = max(hough_space);
[rows, cols] = size(image);
difference = 25;
thresh = max(max(hough_space)) - difference;
for i = 1:size(max_in_col, 2)
if max_in_col(i) > thresh
c(end + 1) = i;
r(end + 1) = row_number(i);
end
end
%% plot all the detected peaks on hough transform image
hold on;
plot(theta(c), rho(r),'rx');
hold off;
%% plot the detected line superimposed on the original image
subplot(1,2,2)
imagesc(image);
colormap(gray);
hold on;
for i = 1:size(c,2)
th = theta(c(i));
rh = rho(r(i));
m = -(cos(th)/sin(th));
b = rh/sin(th);
x = 1:cols;
plot(x, m*x+b);
hold on;
end
Cross Posted:
https://dsp.stackexchange.com/questions/1958/help-understanding-hough-transform
If your regenerated image looks rotated by 90 degrees or otherwise flipped, it's likely because the plotting isn't happening as you expect. You can try axis ij; to move the origin of the plot and/or you can reverse your plot command:
plot(m*x+b, x);
As for peak detection, you might want to look at imregionalmax.