How to project a N*N matrix on a circular shape without loosing any data?

Dear community;

I have a 60 by 60 matrix that represents a temperature profile over a circular surface.

I need to represent the temperature distribution on a circular shape rather show on a rectangular shape.

In time I use imagsc but I am not able to reproduce a circular (heat map like) plot.

Answer 1

[Walter Roberson]

See

https://www.mathworks.com/matlabcentral/fileexchange/17933-polar-to-from-rectangular-transform-of-images?s_tid=srchtitle

https://www.mathworks.com/matlabcentral/fileexchange/49040-pcolor-in-polar-coordinates?s_tid=srchtitle

[Ali Hariri]

Thank you I have already used the Pcolor function that was not the best solution. However thank you so much.

[Walter Roberson]

How did the polor to rectangular contribution work out for you?

[Ali Hariri]

Actually I am now able to reproduce a polar circle shape to report a heat map's like output. Thank you so much.

Answer 2

[Mathieu NOE]

hello

maybe this ?

% your  temperature data
N = 60;
temp_data = 50+10*peaks(N);

% Generate basic cylinder
t = (0:2*pi:2*pi*(N-1))/N;
% r = 2 + 0*sin(t/2); % straigth cylinder
r = 2 + sin(t/2); % inflated cylinder (tire)

[X,Y,Z] = cylinder(r,N);
surf(X,Y,Z,temp_data)
colormap(jet)
colorbar('vert')

[Mathieu NOE]

If you want to tilt your cylinder , you can use this Fex submission :

Generate cylinder with custom axis - File Exchange - MATLAB Central (mathworks.com)

[Ali Hariri]

Is this possible to show on a circle instead of cylinder? Actually the temperature matrix is calculated based on a rectangular surface gometry. however, now it suppose to show a similar area circle.

[Mathieu NOE]

hello again

as suggested by @Walter Roberson, there is already a solution for that

pcolor in polar coordinates - File Exchange - MATLAB Central (mathworks.com)

but it seems you are not yet satisfied with it, so what are you looking for ? showyour results and what you want

seems to me it's working quite well :

% your  temperature data
N = 60;
temp_data = 50+10*peaks(N);

R = linspace(5,10,N); % 
Az = linspace(0,360,N); % in degrees
figure
polarPcolor(R,Az,temp_data,'colormap','jet'); % colormap is here an optional argument

[Ali Hariri]

Thank you so much. I have used both recommendations by you and @Walter Roberson. Initially, I made a mistake, then I see your reccomentatios were actully works.

Thank you so much

[Mathieu NOE]

ok, so you should accept Walter's answer