【作者】 马庭元；

【导师】 张大坤；

【作者基本信息】 沈阳工业大学 ， 计算机应用技术， 2005， 硕士

【摘要】 科学计算可视化(Visualization in Scientific Computing)是发达国家20世纪80年代后期提出并发展起来的一个新研究领域。它运用计算机图形学和图像处理技术,将科学计算过程中产生的数据转换为图形及图像在屏幕上显示出来并进行交互处理。 科学计算可视化技术的核心是三维空间数据场的可视化,它有几种不同的可视化算法。直接体绘制是近年来迅速发展起来的一种三维数据场可视化方法,是当前可视化研究的热点。因为体绘制技术不是对数据场中特定的等值面进行绘制,而是将整个数据场作为整体进行可视化显示,所以能够解释数据场中更加丰富的信息和规律。然而由于体绘制的成像速度较慢,难以达到实时绘制的要求,所以有许多问题需要进行深入的研究。 本文首先对医学三维图像重建中应用的科学计算可视化理论、方法进行了探讨和研究;分类阐述了医学三维图像重建的几种方法,分析了它们的差异和特点;介绍了直接体绘制的几种方法;研究了体绘制实现的光照模型、原理及流程。第二,详细分析了DICOM文件格式。仔细地研究了医学断层图像的体视化的各个过程;对得到的断层图像进行了预处理,并提出了行之有效的解决方案,对数据格式进行转换,建立了自己的体数据结构,便于进行后处理。经过图像转换大大减少了操作的数据量。第三,分析和实现了光线投射体绘制技术的原理和所涉及的光照模型、伪彩色、重采样、图像合成等技术;在重采样中采用包围盒的方法减少了重采样的数量,使处理的数据量大大减少,提高了显示速度。第四,实现了医学头部CT(Computerized tomograpby)断层图像的三维数据场的可视化。观察平面垂直于Z轴,收到了良好的效果。第五,采用数据源为512×512×115的数据集,在微机的环境下,实现了头部CT断层图像的直接体绘制重建。重建后的结果图像比较逼真地表现了重建组织,得到了较为满意的结果。更多还原

【Abstract】 Visualization in Scientific Computing is a new research that the developed countries put forward to the late 80s’ in the 20 century. It applys computer graphics, image processing technique to transform the resulting data from the process of scientific computing to graphics and image that will be display in computer screen.Visualization in Scientific Computing’s core is 3D data field’s Visualization. It has several algorithms. The direct volume rendering is a rapidly developed visualization method of 3D data field in recent years and is a hot research. Because it not only designs the iso-surface of 3D data field but also make the data field as the whole to display. So the direct volume rendering can explain the more rich information and regulation. But the designed image is rather slow and hard to arrive at a real time design. So we necessary further research the thesis to improve the quality and the efficiency of the volume rendering.In this thesis, theory and method about visualization in scientific computing are introduced and described. Several rendering methods and their characters are analyzed. Some methods of the volume rendering and the illumination modal, theory, and the flow are introduced. Second, we exert a group of CT(Computerized tomography) data based on DICOM Principle and detailedly analyzed the volume data’s visualization of the medical and convert the file into bmp. It adduces the data that will be dealed with. Third, we pretreat the image so that it facilites to the next operations. Ray-casting volume rendering and some involved technologies, such as shading model, resampling and mapping are introduced. Forth, I accomplished the visualization of the 3D data field of head CT (Computerized tomography) image.At last, we implement the 3D data field visualization of head CT image. The task’s data resource’s size is 512×512×115,we implemented the reconstruction of head CT . The reconstructed images vividly acquitted the organize, we obtained the anticipate result.更多还原