Accelerating Image Pyramid on GPUs

Article Preview

Abstract:

Image pyramid is an important and essential step in many digital image processing applications. In this paper, we demonstrate that modern GPUs can significantly accelerate image pyramid task by using texture memory and well configuring work group dimension and size. Our two pass method behaves faster when compared with OpenCV's image pyramid GPU implementation. Fast image pyramid on GPUs expects to enable real-time image processing.

You might also be interested in these eBooks

Info:

Periodical:

Advanced Materials Research (Volumes 926-930)

Pages:

403-406

Citation:

Online since:

May 2014

Authors:

Keywords:

Export:

Price:

Permissions CCC:

Permissions PLS:

Сopyright:

© 2014 Trans Tech Publications Ltd. All Rights Reserved

Share:

Citation:

* - Corresponding Author

[1] E. H. Adelson, C. H. Anderson, J. R. Bergen, P. J. Burt, and J. M. Ogden: RCA engineer, Vol. 29 (1984), No. 6, p.33–41.

Google Scholar

[2] Information on http: /www. cse. yorku. ca/ kosta/CompVis Notes/gaussian pyramid. pdf.

Google Scholar

[3] T Lindeberg: Scale-Space Theory in Computer Vision (Kluwer Academic Publishers, UAS 1994).

Google Scholar

[4] M. Macedonia: Computer, Vol. 36 (2003), No. 10, p.106–108.

Google Scholar

[5] Information on http: /www. khronos. org/registry/cl/specs/opencl-1. 2. pdf.

Google Scholar

[6] Information on http: /developer. amd. com/download/AMD/Accelerated Parallel Processing OpenCL Programming Guide. pdf.

Google Scholar

[7] B. G. S. Ramesh Jain, Rangachar Kasturi: Machine Vision. McGraw-Hill, (1995).

Google Scholar

[8] Information on: http: /developer. amd. com/resources/documentation-articles.

Google Scholar