Intuitive volume exploration through spherical self-organizing map

Abstract

Direct Volume Rendering is one of the most popular volume exploration methods, where the data values are mapped to optical properties through a Transfer Function (TF). However, designing an appropriate TF is a complex task for the end user, who may not be an expert in visualization techniques. The Self-Organizing Map (SOM) is a perfect tool to hide irrelevant TF parameters and, through unsupervised clustering, present a visual form of the topological relations among the clusters. This paper introduces a novel volume exploration technique which utilizes the cluster visualization ability of SOM to present a simple intuitive interface to the user for generating suitable TFs. Rather than manipulating TF or cluster parameters, the user interacts with the spherical lattice of the SOM to discover interesting regions in the volume quickly and intuitively. The GPU implementation provides real-time volume rendering and fast interaction. Experimental results on several datasets show the effectiveness of our proposed method. © 2013 Springer-Verlag.