Medical image compression by "neural-gas" network and principal component analysis

This paper presents a new compression scheme for digital still images, by using the "neural-gas" network for codebook design, and several linear and nonlinear principal component methods as a preprocessing technique. We investigate the performance of the compression scheme depending on the blocksize, codebook and number of chosen principal components. The nonlinear principal component method shows the best compression results in combination with the "neural-gas" network.