Implementation of Parallel Fast Hartley Transform (FHT) Using Cuda

Hovhannes Bantikyan^1,

¹Department of Computer Systems and Informatics, State Engineering University of Armenia, Yerevan, Armenia

Cite this paper:
Hovhannes Bantikyan. Implementation of Parallel Fast Hartley Transform (FHT) Using Cuda. Journal of Computer Sciences and Applications. 2014; 2(1):6-8. doi: 10.12691/jcsa-2-1-2

Abstract

Implementation of Fast Hartley Transform in parallel manner on Graphics Processing Unit, using CUDA technology is presented in this paper. Calculating FHT in parallel, using multiple threads, gives us huge improvement in calculation speed. Developed CUDA based parallel algorithm, which experimental results compared with results of CPU based sequential algorithm. Edge detection algorithms can be speed up for large images, performing in frequency domain. Here experiments are done on various edge detection filters and different image sizes, using fast Hartlay transform.

Keywords:
Fast Hartley Transformation parallel computing GPGPU CUDA programming

This work is licensed under a Creative Commons Attribution 4.0 International License. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/

Figures

References:

[1]	Bracewell R. N., The Hartley Transform, Oxford, Oxford Univ. Press, 1986, 168 pages.
[2]	Scott R., Doing Hartley smartly, EE Times-India, 2000.
[3]	Millane R. R., Analytic Properties of Hartley Transform and their Implications, IEEE, 1994.
[4]	Henning K., Maurico D., and Jurgen F., The Hartley transform in seismic imaging, GEOPHYSICS VOL. 66, NO. 4 (JULY-AUGUST 2001), Pages. 1251-1257.
[5]	Somasundaram M. Implementation and performance evaluation of parallel FFT algorithms.
[6]	B. Jähne., Digital Image Processing. 2005.
[7]	J. Sanders, E. Kandrot., CUDA by Example. 2010.
[8]	NVIDIA CUDA C Programming Guide. NVIDIA Corp. 2012.
[9]	H. Lensch, R. Strzodka., Massively Parallel Computing with Cuda. 2010.