The Key Technology Of Three-dimensional Visualization Of Isosurface Extraction Studies

Iso-surface extraction method is the most commonly used visualization techniques. It has been widely used in medical imaging, surface reconstruction and three-dimensional modeling of geological data etc. However, with the further development of science, requirements are increasingly high on the Iso-surface extraction techniques. Currently, it is still very difficult that extracting interactively Iso-surface from three-dimensional data field. Therefore, it is very meaningful to study the key technology of Iso-surface extraction.In this paper, the data preprocessing, triangulation and tetrahedral mesh technology of Iso-surface extraction methods are researched based on borehole data. The specific contents are as follows:(1)Because K-Means method is sensitive to the isolated points, in this paper it is introduced into the data preprocessing to filter out isolated points. The result of experiments shows that this method can effectively filter out remote isolated points and reduce the impact of the noise data.(2)By analyzing the existing triangulation methods, we discovered that these methods are all based on Delaunay criterion to achieve. Therefore, these algorithms are all restricted by this criterion on implementation. For this case, two new triangulation algorithms are proposed which do not fully meet the Delaunay criteria in this paper. The first triangulation algorithm is the approximation convex hull which basic steps as follows:firstly select the data points on the approximation convex hull. Secondly, scan data points within the approximation convex hull to divide it into a number of triangles or trapezium which have some points on edges. Finally, these triangles or trapezium are further divided into simple triangles. The idea of this method is simple and it is easy to be implemented. However, it can not be applied to the data set with constraint conditions. Therefore, in order to solve this problem, a new algorithm named constraint convex hull triangulation is proposed by improving the approximate convex hull triangulation method. First of all the algorithm divided data into two parts which are data with constraints and data without constraints, and then use the first triangulation algorithm to deal with the two parts respectively.(3) By analyzing the time complexity of proposed algorithms in this paper, the results show that the efficiency of the proposed algorithms is higher than the traditional algorithms. In addition, an analysis of the shortcomings of the algorithms is given, in order to provide a basis for further study.