Abstract: Current acceleration methods are not adequate for the interaction design of transfer functions.In order to resolve the conflict，a new volume rendering acceleration algorithm，the edge cutting acceleration algorithm（ECAA），is proposed．To overcome the two key difficulties，which are，firstly，how to eliminate the dependency on transfer function adjustments and，secondly，how to determine the empty range of voxels，ECAA has established a set of identifying rules for absolute idle voxels and designed an effective separating mechanism for marginal idle voxels.These have structured a new acceleration mode independent of transfer function adjustments.ECAA evidently has upgraded the rendering speed，and at the same time，kept a high rendering quality.The edge cutting procedure of ECAA is in the pre-processing phase and its parameter selection is very easy.So ECAA is of extensive adaptability and very appropriate to the applications of popular medical image analysis systems，which demands interacted transfer function adjustments.With normal cutting mode of ECAA，in next steps，shrunk volume data are highly convenient for using ray casting or splatting algorithms and combining with other acceleration techniques，so as to reach an integration of acceleration effects with different characteristics.ECAA is a good complement to other acceleration techniques.Different practical examples have presented the validity and advantages of this algorithm.

摘要： 针对目前加速方式与传递函数交互设定需求的矛盾，提出了一个新的基于边缘切除原理的体绘制加速算法。算法针对两个关键难点：如何消除传递函数调整依赖性，如何识别空体素，提出了有效的绝对空体素识别准则，设计了高效的边缘空体素分离机制，构成了不依赖传递函数调整的加速模式。在保持高的图像质量的前提下，边缘切除算法具有显著的绘制速度提升。边缘切除过程在预处理阶段进行，算法参数易于选取和推广，具有广泛的适应性，非常适合需要交互设定传递函数的普及型医学图像分析系统应用。算法采用了规则的边缘切除方式，收缩后的体数据非常方便后续光线投射或溅射算法应用，可以方便地与其他各种加速方式组合使用，使不同角度的加速效果实现叠加，是当前各种主流加速技术的一个很好的互补技术。不同背景的运算实例，测试和验证了算法的有效性。