The Volumetric Clipping for Industrial Computerized Tomography Based on CUDA

Article Preview

Abstract:

The fast development of the Graphics Processing Unit’s progammability and parallel computing power provides new solutions for visualization technology. As the carrier of volume date, a cubic box is selected as the bounding box in the improved ray-casting algorithm, whose diagonal vertex coordinates are stored in boxMin and boxMax respectively. By adjusting the two parameters we can change the shape of the bounding box and then choose the interesting region of the object, a CUDA based interactive volumetric clipping tool is finally realised. By changing the shape of bounding box, the axial volumetric clipping and arbitrary cross-section clipping are important ways to help to get the object's internal information. Experiments show this method is simple, efficient and useful especially for the irregular shape defect. In addition, even for the larger volume of data, we can get a high quality, cross-sectional images on a fast speed.

You might also be interested in these eBooks

Info:

Periodical:

Pages:

324-327

Citation:

Online since:

December 2012

Export:

Price:

Permissions CCC:

Permissions PLS:

Сopyright:

© 2013 Trans Tech Publications Ltd. All Rights Reserved

Share:

Citation:

[1] J. Sanders and Edward: CUDA by Example (China Machine Press, Beijing 2011).

Google Scholar

[2] Nvidia: CUDA Compute Unified Device Architecture: Progamming Guide Version 4. 0.

Google Scholar

[3] M. Levoy: Computer Graphics and Applications, IEEE, Vol. 8 (3) (1988), pp.29-37.

Google Scholar

[4] CUDA Volume Rendering, Information on http: /blog. sina. com. cn/s/ blog_4addf54b0100ofo6. ht ml.

Google Scholar

[5] Y. Z. Kang: GPU programming and CG language (Banshan studio, China, 2009) , In Chinese.

Google Scholar

[6] M. Hadwiger and P. Ljung: ACM SIGGRAPH, (2009), pp.2-53.

Google Scholar

[7] K. Engel and M. Hadwiger: New York AK Peters, (2006), pp.381-413.

Google Scholar