Accelerating Model Modification and Output Triangle Mesh by Using GPU in NURBS-base CAD System

• Main Authors: Chien-ting Yeh, 葉建廷
• Other Authors: Hong-tzong Yau
• Format: Others
• Language: zh-TW
• Published: 2008
• Online Access: http://ndltd.ncl.edu.tw/handle/07353072082159969948

碩士 === 國立中正大學 === 機械工程所 === 96 === Nowadays, CAD system is the favorite design tool. The traditional CAD system is based on freeform surface model, and the designer devises and edits the surface model. Finally the designer outputs the information to manufacture the product. Although, freeform surface is able to describe objects’ geometry exactly but inefficiently due to necessarily of complicated non-linear computation in various applications. As a result, such system usually becomes time-consuming and unstable. In addition, STL model is more and more popularly using in manufacture system. There are many researcher to study intensively in that, which is surface model translate to triangle mesh model. However, in this section, it also wastes a lot of time to operator the result. Therefore, if we can combine with their advantages and overcome the problem of time-consuming and unstable. The system must be an important tool in CAD system. In order to decrease the operator time, we use graphic card to procure this purpose. Graphic Processor Unit(GPU) is a type of CPU attached to a Graphics card dedicated to calculating floating point operations and GPU is that of stream processing. This concept uses massively parallel floating-point, yet dedicated computational power of a modern graphics accelerator''s shader pipeline. Therefore, we propose to use GPU to accelerate model modification and render.