Precission of Iso-Surface Extraction from Volume Data and Visualization

Abstract

There are various methods for extracting iso-surfaces from volumetric data. Many specific techniques have been developed to increase the speed of computation and decrease memory requirements. A comparison of the selected methods was made in different aspects. Surprisingly, experiments proved, that there is no direct relation between precision and human perception of the extracted iso-surface. Although a precision of the iso-surface extraction is very important as well, it is not mentioned usually. After the extraction, the surface (usually the triangular mesh) is displayed by a standard package that uses algorithms like constant, Gouraud or Phong shading, for shading polygonal meshes. Unfortunately the Gouraud and Phong shadings cannot be used in some cases as these techniques prefer the "smoothness" to "precision" of the displayed surface. In cases, when the colour of the triangle of the triangular mesh is given in a representative inner point for each triangle, both methods can lead to very misleading results. The S-shading method that respects the smoothness and precision in some extent can be used.