You can normalise the values in the matrix so that they lie between 0 and 1 and then use the command imshow
:
// create a random example of a matrix
I = 4000*rand(256, 256, 3);
// normalise the values in I
for i = 1:3
I(:, :, i) = I(:, :, i)/max(max(I(:, :, i)));
end
// display as image
imshow(I, 'InitialMagnification', 'fit')