A self-organizing algorithm for image compression

Presents a modification of Kohonen´s algorithm used in designing codebooks for vector quantization (VQ) of images. Kohonen´s original algorithm builds up a map of the input signal in a one or two dimensional array of neurons. In the present work, the map is built in the synaptic space itself. Another modification is introduced: instead of finding the winning neuron around which the neighborhood is defined, a k-dimensional sphere (neighborhood) is centered at the training vector itself, representing thus a great simplification in the original algorithm. Simulation results show that the proposed method performs better than the traditional LBG algorithm for all tested image, at all bit per pixel rates evaluated