基于梯度特征的医学体绘制传输函数曲线拟合方法

Curve Fitting Method for Transfer Functions Based on Gradient Characteristics for Medical Volume Rendering

医学体绘制的精度和实时性取决于传输函数曲线的精度和生成过程的运算开销。典型体绘制的传输函数生成方法是通过灰度映射来生成分段线性函数,精度低、耗时长。本文针对该问题,提出了一种基于体数据梯度特征的传输函数曲线拟合方法。该方法根据体数据集的梯度变化特点,利用数据值、梯度和沿梯度方向的二阶导数拟合曲线函数来生成传输函数。实验结果表明:在使用规则体数据时,本方法可有效实现医学可视化。

In medical volume rendering, the precision and the real-time rest with the precision and computation are determined by expense of the generation process for transfer function curve. The typical transfer function has a low precision and is time-consuming, thus it is generated by gray mapping to the segment linear function. This paper proposes a new method based on the gradient characteristics of volume data. The data value and its first and second directional derivatives are used along the gradient direction to synthesize a curve function aimed at regular volumetric data. The experimental result shows that the method is applicable in medical visualization.