真实感图形显示系统开发环境的研究

A Development Testbed for Realistic Image Generation

介绍一个用于真实感图形研究的开发环境——TESTRIG系统的结构及基本原理.该系统能够测试各种不同的新算法,适合许多用户的需要.而且具有友好的用户界面及多种测试工具,用户易学易用,因此该系统为研究真实感图形显示理论提供了一个良好的开发环境.它的模块性结构及各模块之间的友好接口,使系统具有很好的灵活性.此外还提出了一种基于几何信息的反走样算法及快速扫描线算法.

It appears inadvisable to build up a complete graphics display system with much effort just in order to test the algorithm which is developed by one researcher. Therefore, it is very necessary to create a graphics display system for realistic image generation. This paper describes such a graphics display system or testbed for realistic image generation which is named as TESTRIG (Test for Realistic Image Generation). TESTRIG has following characteristics: 1. TESTRIG can test various new algorithms and can satisfy the individual requirements of each user. TESTRIG provides multiple input routines in environment modeling. Various geometrical models the users need can be easily built up by TESTRIG in ns of input information. TESTRIG also have the capabilities of editing each geometrical element of the model. Three famous illumination models which were presented by Phong, Cook and Torrance [3], and Glassner[6] respectively are available in TESTRIG. In the generation of the realisitic images, TESTRIG provides three typical algorithms-Fast Scan Line, uaive ray tracing and Glassner's ray tracing. The user can easily replace any block of the system by his own block. 2. TESTRIG provides researchers with a friendly user interface. TESTRIG is an open system. Not only can the users call various powerful subroutines directly, but also modify some important source files of the system. Many test tools for routines are also available. 3. TESTRIG is easy to learn and to use. For a new or inexperienced user, TESTRIG is a general realistic image display system. If the user forgets to provide TESTRIG with information, it can ask for such information to be provided. Many flexible permanent menus, pull down menus and popup menus are also available. In addition to the characteristics mentioned above, two new algorithms developed by the authors are used in TESTRIG. One is ABGI (Antialiasing Based Geometry Information), which is very efficient for the aliasing on the boundaries of images. The other is FSL (Fast Scan Line). Our statistics show that FSL is twice as fast as Watkins algorithm.