Creating an RGB Matrix from a 1 by N vector in Matlab

Question

I have a vector of N numbers ranging in values from 0 to 1. Is there a way of getting Matlab to create a 3 by N matrix of RGB values depending on the value of the index of the initial vector?

I would like the colours to range from blue for 0 to red for 1, but go through intermediate colours as in the jet colourmap. So far, I have successfully got Matlab to range from blue to red, but with purple elements in between.

Any help would be greatly appreciated.

Thanks, Ryan

Code so far:

%Adapt colour values so that they are between 0 and 1
ixcolours=(Ix(:,1)+abs(min(Ix(:,1))))/max(Ix(:,1)+abs(min(Ix(:,1))));
iycolours=(Iy(:,1)+abs(min(Iy(:,1))))/max(Iy(:,1)+abs(min(Iy(:,1))));

for i=1:471;
    if seed_locs(i,3)==20; %If we are in the 20th seed z slice
        plot(seed_locs(i,1),seed_locs(i,2),'MarkerFaceColor',[ixcolours(i) 0 1-ixcolours(i)],'MarkerEdgeColor',[ixcolours(i) 0 1-ixcolours(i)],'MarkerSize',10,'Marker','s') %plot the x and y seedlocs
        hold on
    elseif test_locs(i,3)==20; %if we are in the 20th test z slice
        plot(test_locs(i,1),test_locs(i,2),'MarkerFaceColor',[iycolours(i) 0 1-iycolours(i)],'MarkerEdgeColor',[iycolours(i) 0 1-iycolours(i)],'MarkerSize',10,'Marker','s') %plot the x and y seedlocs
    end
end

Update

Thanks for the help lhcgeneva and FirefoxMetzger, both of your algorithms are useful.

I have tried implementing FirefoxMetzger's code, but now I'm presented with an issue regarding imagesc. Before, my full code was:

 %Create Figure with handle.
h5=figure('units','normalized','outerposition',[0 0 1 1]);
whitebg(h5,[0 0 0]);
subplot(2,5,1);
k=1;
for i=16:25
    subplot(2,5,k);
    imagesc(squeeze(ana(:,:,i)));
    title(['Z=',num2str(i)]);
    hold on
    colormap gray
    axis equal
    k=k+1;
end
%Adapt colour values so that they are between 0 and 1. We want to scale
%both data sets equally, so we find the smallest value across Ix and Iy. We
%also find what will be the new largest value across Ix and Iy, after we
%add the magnitude of the smallest value to make all numbers greater than
%or equal to 0.
absolutemin=min(min(Ix(:,1)),min(Iy(:,1)));
absolutemax=max(abs(absolutemin)+(max(Ix(:,1))),abs(absolutemin)+max(Iy(:,1)));

%Add the smallest value, and divide by the largest maximum value for both Ix
%and Iy.
ixcolours=(Ix(:,1)+abs(absolutemin))/absolutemax;
iycolours=(Iy(:,1)+abs(absolutemin))/absolutemax;

o=1;
for k=16:25; %For all 3D slices
    for i=1:471; %and for all x and y seed slices
        if k==seed_locs(i,3);
            subplot(2,5,o); %go to the corresponding z subplot
            plot(seed_locs(i,1),seed_locs(i,2),'MarkerFaceColor',[ixcolours(i) 0 1-ixcolours(i)],'MarkerEdgeColor',[ixcolours(i) 0 1-ixcolours(i)],'MarkerSize',10,'Marker','s') %plot the x and y seedlocs
            hold on
        end
    end

    for i=1:486;
        if k==test_locs(i,3);
            subplot(2,5,o);
            plot(test_locs(i,1),test_locs(i,2),'MarkerFaceColor',[iycolours(i) 0 1-iycolours(i)],'MarkerEdgeColor',[iycolours(i) 0 1-iycolours(i)],'MarkerSize',10,'Marker','s') %plot the x and y seedlocs
        end

    end
    o=o+1; %go to the next z subplot
end

where I would have a 2 by 5 subplot of a brain with data superimposed on top of it as squares through the plot command.

However, now if I try to use the RGB values, I either only get images of the brain, or just the plotted points, but no superimposing.

 %Adapt colour values so that they are between 0 and 1. We want to scale
%both data sets equally, so we find the smallest value across Ix and Iy. We
%also find what will be the new largest value across Ix and Iy, after we
%add the magnitude of the smallest value to make all numbers greater than
%or equal to 0.
absolutemin=min(min(Ix(:,1)),min(Iy(:,1)));
absolutemax=max(abs(absolutemin)+(max(Ix(:,1))),abs(absolutemin)+max(Iy(:,1)));

%Add the smallest value, and divide by the largest maximum value for both Ix
%and Iy.
ixcolours=((Ix(:,1)+abs(absolutemin))+1/64)/(absolutemax+1/64);
iycolours=((Iy(:,1)+abs(absolutemin))+1/64)/(absolutemax+1/64);
h1x=figure();
colour_values_x=interp1((1:64)/64,colormap(h1x,'jet'),ixcolours);
close(h1x)
h1y=figure();
colour_values_y=interp1((1:64)/64,colormap(h1y,'jet'),iycolours);
close(h1y)

%Create Figure with handle.
h5=figure('units','normalized','outerposition',[0 0 1 1]);
whitebg(h5,[0 0 0]);
subplot(2,5,1);
k=1;
for i=16:25
    subplot(2,5,k);
    imagesc(squeeze(ana(:,:,i)));
    title(['Z=',num2str(i)]);
    colormap gray
    axis equal
    k=k+1;
end
figure(h5);
o=1;
for k=16:25; %For all 3D slices
    for i=1:471; %and for all x and y seed slices
        if k==seed_locs(i,3);
            subplot(2,5,o); %go to the corresponding z subplot
            plot(seed_locs(i,1),seed_locs(i,2),'MarkerFaceColor',[colour_values_x(i,1) colour_values_x(i,2) colour_values_x(i,3)],'MarkerEdgeColor',[colour_values_x(i,1) colour_values_x(i,2) colour_values_x(i,3)],'MarkerSize',10,'Marker','s') %plot the x and y seedlocs
            hold on
        end
    end

    for i=1:486;
        if k==test_locs(i,3);
            subplot(2,5,o);
            plot(test_locs(i,1),test_locs(i,2),'MarkerFaceColor',[colour_values_y(i,1) colour_values_y(i,2) colour_values_y(i,3)],'MarkerEdgeColor',[colour_values_y(i,1) colour_values_y(i,2) colour_values_y(i,3)],'MarkerSize',10,'Marker','s') %plot the x and y seedlocs
        end

    end
    o=o+1; %go to the next z subplot
end

lhcgeneva lhcgeneva · Accepted Answer · 2015-11-24T22:36:11

MATLAB has a built-in for this. If N is the number of shades you want (471 in your case) you can do

N = 471;
c = colormap(jet(N));

then access c in every loop iteration via c(i, :). For reference see the man page ;).

Creating an RGB Matrix from a 1 by N vector in Matlab

Update

2 Answers