Constraining global interpolation methodsfor sparse data volume visualization

The study of sparse data volume visualization further expands the application scope of volume visualization from scientific study and medical imaging into engineering applications. The existing two-step approach to sparse data volume visualization faces three dilemmas: the negative-value dilemma, the zero-value dilemma, and the correctness dilemma. This paper presents three constraining techniques that can be used to alleviate these dilemmas. The constraints are point constraint, value constraint, and local constraint. By adding information that is not explicitly carried by the sample data, the constraining techniques guide the global interpolation methods to avoid misrepresenting the sample data. Examples are given to demonstrate the use of the techniques in visualizing sparse data from environmental engineering applications.