基于直方图熵的体数据分类算法研究

体数据的分类用于确定体素的可见性，在三维体绘制中起着重要的作用。提出一种基于熵的体数据分类算法。首先根据累计直方图将体数据的直方图进行分段，然后根据熵判别式在每个分段中计算一个阈值作为阻光度传递函数的分段点，再根据阻光度传递函数计算出体素的阻光度值，完成体数据的分类。以工业CT体数据为对象进行实验，其结果表明，所提出的算法较好地实现了体数据的分类，体绘制结果清晰，且能够实现试件的模拟拆卸。

Classification of volume data based on entropy of histogram

The classification step used to assign the appropriate opacity to each voxel is very important in the volume rende-ring.This paper proposed a new classification algorithm for volume data,which was based on the entropy.Firstly,partitioned the histogram of the volume data into different subsections through calculating the accumulated histogram of volume data according to the number of object classes.Secondly,in each subsection computed a threshold based on the entropy.Finally,assigned the opacity of voxel by a transfer function,which was split into subsection by these thresholds.Experimental results from industrial CT images were presented and it indicated that the classification of volume data was obtained with the better accuracy by the proposed algorithm.The industrial components are shown explicitly in the volume rendering results and successfully disassembled in the computer simulated.