Fake colorbar for image

I would like to be able to show an image like the example below, but instead make a fake color scale that is not related to the image values. The scale would ideally look like a colorbar, but I would choose the number of colors in the scale, the text label of each color, and its RGB value. My example below might have 4 color patches: 'Red Pepper' [194 24 25], 'Green Pepper' [157 153 47], 'Purple Cloth' [134 94 157], and 'White Garlic' [255 255 255]. Or it could have a different number of color patches. I do not want to do any image processing on the image, just annotate it. Is this possible?

figure;
I = imread('peppers.png'); % Read in example image
imshow(I); % Show image
colorbar; % Not what we want, since it is a continuous color map which is automatically labelled 0, 0.2, 0.4, ...1 
% Instead we want a color scale with with the color patch RGB values and text labels supplied by the user

Answer 1

[Image Analyst]

Do you mean like this:

% Read in the color demo image.
[rgbImage, colorMap] = imread('peppers.png');
imshow(rgbImage, colorMap);
axis on;
title('Original Color Image', 'FontSize', fontSize);
% Enlarge figure to full screen.
set(gcf, 'Units', 'Normalized', 'OuterPosition', [0 0 1 1]);
drawnow;
% 'Red Pepper' [194 24 25], 'Green Pepper' [114 24 55], 'Purple Cloth' [117 74 26], and 'White Garlic' [255 255 255].
cMap = [194 24 25;114 24 55; 117 74 26; 255 255 255]/255
colormap(cMap);
colorbar('Ticks',32:64:255,...
  'TickLabels',{'Red Pepper','Green Pepper','Purple Cloth','White Garlic'},...
  'FontSize', 20);

I think your color definitions are strange though. Purple and green seem switched.

[KAE]

Yes, my mistake. Now corrected in the example, sorry!

Answer 2

[Rik]

As long as you're not using the colormap to show anything, you can use the method below. If you need to use the colormap, you can use ind2rgb.

I = imread('peppers.png'); % Read in example image
imshow(I); % Show image
h=colorbar;
cmap=[194 24 25;114 24 55;117 74 26;255 255 255];
cmap=cmap/255;
TickLabels={'Red Pepper','Green Pepper','Purple Cloth','White Garlic'};
Ticks=linspace(0,1,numel(TickLabels)+1);Ticks=Ticks(2:end)-diff(Ticks)/2;
colormap(cmap)
h.TickLabels=TickLabels;
h.Ticks=Ticks;

[KAE]

This works great, as does the other answer. I wasn't sure what to do so I accepted the one with the screenshot.