Polarlicht 2 kmeans 16 large

File:Polarlicht 2.jpg reduced to 16 colors using the k-means algorithm. Implemented in MATLAB as follows:

1. Read in source image.
2. Convert to 3-dimensional matrix of [y-value, x-value, rgb] using imread. Each entry is an integer between 0 and 255 representing the value of the color indicated by rgb at (x, y).
3. Reduce linear resolution 8 times by averaging the values in each 8x8 matrix block (for faster processing).
4. Run k-means (self-implemented) for 20 iterations with k = 16, obtaining 16 colors which are the centroids of each cluster.
5. For each pixel in the original image, replace its value with the closest color based on Euclidean norm of RGB.