Reduce discretisation in matrix

Hello everybody.

I need your help. I have a matrix representing 200000 points. x, y and z coordinates and a pressure value for each point. I manage to vizualise my points with a colorbar for pressure values using plot3c ( from Uli Theune, University of Alberta) but the generated figure is too heavy and my computer crashes. I would like to reduce the number of points and have less pressure values? I would like to plot the same figure but "pixelized". I hope I make mylself clear. If you need some extra info please don't hesitate.

Thank you!

[Star Strider]

What is the actual size of your matrix?

That is:

sz = size(matrix)

what is sz?

[Jean]

HI Star,

My matrix is 198808*4 (4 columns).

Thaks for you help

[per isakson]

See pcolor, Pseudocolor (checkerboard) plot

[Jean]

Thanks per isakson; I tried to use pcolor but I don't understand how to use it in y case. Could you give me an example please?

Answer 1

[Star Strider]

That is actually easier than I thought it would be. If you want to take every 4th row for instance:

decmat = matrix(1:4:end, :);

or with an example:

X = repmat(1:25, 4, 1)';
Xdec = X(1:4:end,:);

Your ‘decimated’ matrix ‘decmat’ or ‘Xd’ is now ¼ the size of your original matrix. You can choose different intervals or even random rows to put into your ‘decimated’ matrix.

Is this what you want to do?

[Jean]

This is good but not exactly what I am looking for. In this case I just suppress a row every n rows. But I am not doing any kind of average isn't it? I am looking for something that will read the values on 5 adjacent rows and calculate the average for those 5 values and make a new row.

[Star Strider]

You may have to experiment with this with your data, but in this simulation at least, it works and should give you the means of every 5 rows in the ‘Xm’ matrix:

X = repmat(1:30, 4, 1)';            % Created Simulated Data Matrix
Xs = reshape(X, 5, [], 4)           % Reshape To Allow Mean Calculations
Xp = permute(Xs, [1 3 2])           % ‘Transpose’ 3D MAtrix
Xm = squeeze(mean(Xs,1))            % Calculate Means