Comparing three PCA-based methods for the 3D visualization of imaging spectroscopy data

In this paper we compare the quality of three different principle component analysis (PCA) based methods to generate transfer functions for the 3D visualization of imaging spectroscopy data. We discuss three criteria for judging the quality of features in these visualizations. These criteria are used to interpret visualizations of features in the brain of the snail Lymnaea Stagnalis. We show that the PCA method that uses model additional information, clearly results in superior visualizations.