基于GPU光线投射算法的心脏体数据可视化

Visualization of Cardiac Volume Data Based on GPU-based Raycasting Algorithm

针对分割后的心脏体数据显示问题,提出了基于GPU光线投射算法的高质量三维可视化新方法。该方法为了解决传统光线投射算法绘制速度慢的问题,采用GPU进行加速,并结合空体素跳过和光线提前终止忽略无效体素的采样。同时,为了获得高质量的重建结果,该方法结合心脏体数据的特点设计传递函数,基于体素梯度模自适应地调整采样步长,并应用BlinnPhong多光源光照模型增强可视化效果。实验结果表明,该方法在实现实时绘制的同时,能够获得高质量的体绘制效果。

Aiming at the problem of displaying the segmented cardiac volume data,an approach based on GPU-based raycasting algorithm was proposed,which was used for producing high image quality. To improve the rendering speed of traditional raycasting algorithm, employed GPU-accelerated algorithm was combined with the empty voxels skipping and early ray termination for ignoring invalid voxels sampling. Moreover,in order to obtain high-quality rendering results, the transfer function was designed according to cardiac volume data,and the sampling step was adaptively adjusted on base of the gradient. In addition,the Blinn-Phong illumination model with multiple lamp-houses was also employed to enhance the visualization effect. The results showed that the proposed method could meet the requirements of real-time rendering and obtain high-quality volume rendering results.