I have an image in which i need different colours for each regions
for example my original image is
i need as
please help
No products are associated with this question.
Convert the RGB image to indexed. Change the colormap.
Walter i tried
I=imread('imagesh1.bmp');
imshow(I)
[IND,map] = rgb2ind(I,4);
imagesc(IND)
colormap(jet)
bu did not get the figure as i posted above in link,please help
That's because your image has already been pseudocolored. What you did will work but essentially is just random colors. You need the original multispectral images - a set of monochrome images taken at each wavelength. Then see my color classification demos in my File Exchange. But even that won't do it because there is more than just color classification going on, as you can see from a careful inspection of the image. Why don't you replicate the method used in the paper?
You need to do color classification on the image. Look in each spectral band and do whatever you need to do to get a classified image. This could be gray level segmentation, texture segmentation, or some higher level conceptual classification (trees, buildings, etc.). Once you have a gray level image where each number represents a class (e.g. 0=water, 1=forest, 2=road, 3=buildings), then you can apply a colormap to make it more visually distinct. Go here for articles on classification: http://iris.usc.edu/Vision-Notes/bibliography/contentscartog.html#Cartography,%20Aerial%20Images,%20Remote%20Sensing,%20Buildings,%20Roads,%20Terrain,%20ATR
0 Comments