In contrast to previous work which focused on the viewpoint selection or illumination optimization separately, this paper proposes the collaborative selection of the viewpoint and lighting parameters. With the given transfer function, the collaborative optimization employs a viewpoint and lighting evaluation framework, which is based on entropy of information and consists of viewpoint entropy, light entropy and the coupling entropy, to search the best viewpoint and lighting parameters to reveal the maximum information about features inside the volume data. The viewpoint entropy evaluates the distribution of opacity; the light entropy takes into account the distribution of colors in the rendered image segmented by the mean shift algorithm; the coupling entropy measures the additional amount of information brought about by the brightness difference between unilluminated and illuminated images. Our experiments with several volume data sets demonstrate the effectiveness of the proposed method.%针对体绘制中只对视点或光照参数的单独优化而导致体数据中隐含的信息无法在图像空间中表现最大化的问题,提出一种视点和光源位置协同优化的方法,在传输函数不变的情况下选择最优的视点和光照参数.该方法基于信息熵设计,并南视点熵、光照熵及两者之间的互偶熵组成视点与光照统一评估框架,用来评估不同视点和光照参数条件下最终绘制图像表达的信息量.视点熵用于评估与视点相关的不透明度值分布所包含的信息量;光照熵的计算则通过考虑颜色在图像空间的分布,使用基于区域的mean shift算法进行基于颜色的图像空间划分;互偶熵用于评估光照前后绘制图像亮度变化所带来的信息量.对各种体数据的实验结果表明,采用文中方法能够有效地协同优化视点和光源位置.
展开▼
