Svakhine Nikolai A, Ebert David S, Andrews William M
Adobe Systems Inc., San Jose, CA 95110, USA.
IEEE Trans Vis Comput Graph. 2009 Jan-Feb;15(1):77-86. doi: 10.1109/TVCG.2008.56.
Volume illustration can be used to provide insight into source data from CT/MRI scanners in much the same way as medical illustration depicts the important details of anatomical structures. As such, proven techniques used in medical illustration should be transferable to volume illustration, providing scientists with new tools to visualize their data. In recent years, a number of techniques have been developed to enhance the rendering pipeline and create illustrative effects similar to the ones found in medical textbooks and surgery manuals. Such effects usually highlight important features of the subject while subjugating its context and providing depth cues for correct perception. Inspired by traditional visual and line-drawing techniques found in medical illustration, we have developed a collection of fast algorithms for more effective emphasis/de-emphasis of data as well as conveyance of spatial relationships. Our techniques utilize effective outlining techniques and selective depth enhancement to provide perceptual cues of object importance as well as spatial relationships in volumetric datasets. Moreover, we have used illustration principles to effectively combine and adapt basic techniques so that they work together to provide consistent visual information and a uniform style.
体数据可视化可以用来深入了解来自CT/MRI扫描仪的源数据,其方式与医学插图描绘解剖结构的重要细节非常相似。因此,医学插图中使用的经过验证的技术应该可以应用于体数据可视化,为科学家提供可视化数据的新工具。近年来,已经开发了许多技术来增强渲染流程,并创建类似于医学教科书和手术手册中的插图效果。此类效果通常会突出主体的重要特征,同时弱化其背景,并提供深度线索以实现正确感知。受医学插图中传统视觉和线条绘制技术的启发,我们开发了一组快速算法,用于更有效地强调/弱化数据以及传达空间关系。我们的技术利用有效的轮廓勾勒技术和选择性深度增强,以提供体数据集中对象重要性以及空间关系的感知线索。此外,我们运用插图原理有效地组合和调整基本技术,使它们协同工作,以提供一致的视觉信息和统一的风格。