中文摘要：

传统体绘制方法难以清晰地展现体数据的内部结构, 本文提出了基于连续框架的利用数据拓扑特征增强体绘制的新方法. 该方法首先利用7 方向box 样条拟插值对离散数据进行连续重建, 然后通过求解多项式系统得到数据场中的特征点分布; 从鞍点出发可以在连续模型上计算出鞍极曲线, 本文提出以鞍极曲线长度给特征点值分布加权得到含权特征点直方图, 并依据直方图信息重新设计体绘制的传输函数来增强体绘制. 实验结果表明, 与离散框架的方法相比, 文中方法更加简单可靠, 且可视化结果具有更好的平滑性, 能清晰地反映体数据内部的细微结构.

英文摘要：

Traditional visualization approaches for volumetric data are not able to reflect clear interior structures of input data. We present a method to enhance volume rendering by topological features under a continuous framework. A continuous field is first reconstructed from discrete data by 7-directional box spline quasi-interpolation, and critical points are obtained from gradient polynomial systems. Then sad-dle-extremum arcs are computed from the continuous field, and the lengths of them are used to build a weighted critical value histogram, which helps to design a new transfer function. Compared with those ex-isting discrete approaches, our method is easier to implement and the results are smoother and clearer.