Multicore implementation of the windowed Fourier transform algorithms for fringe pattern analysis

Ming Zhao; Qian Kemao

doi:10.1364/AO.54.000587

Applied Optics
Vol. 54,
Issue 3,
pp. 587-594
(2015)
•https://doi.org/10.1364/AO.54.000587

Multicore implementation of the windowed Fourier transform algorithms for fringe pattern analysis

Ming Zhao and Qian Kemao

Not Accessible

Your library or personal account may give you access

Get PDF
Email
Share
Get Citation
Copy Citation Text
Ming Zhao and Qian Kemao, "Multicore implementation of the windowed Fourier transform algorithms for fringe pattern analysis," Appl. Opt. 54, 587-594 (2015)

Export Citation
- BibTex
- Endnote (RIS)
- HTML
- Plain Text
Citation alert
Save article

Abstract

Due to its effectiveness in fringe pattern analysis, a fast implementation of windowed Fourier transform (WFT)-based algorithms is desired. In this work, we speed up the implementation in two aspects. First, we speed up the computation of Fourier transform, which is a core component of the WFT-based algorithms. Second, we parallelize the main body of these algorithms due to its parallel characteristic. With these two approaches, we obtain a $10 \times$ faster implementation than the original MATLAB version.

Full Article | PDF Article

More Like This

Real-time 2D parallel windowed Fourier transform for fringe pattern analysis using Graphics Processing Unit

Wenjing Gao, Nguyen Thi Thanh Huyen, Ho Sy Loi, and Qian Kemao
Opt. Express 17(25) 23147-23152 (2009)

Windowed Fourier transform for fringe pattern analysis: theoretical analyses

Qian Kemao, Haixia Wang, and Wenjing Gao
Appl. Opt. 47(29) 5408-5419 (2008)

Windowed Fourier transform for fringe pattern analysis

Qian Kemao
Appl. Opt. 43(13) 2695-2702 (2004)

Cited By

You do not have subscription access to this journal. Cited by links are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access Optica Member Subscription

Figures (2)

You do not have subscription access to this journal. Figure files are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access Optica Member Subscription

Tables (6)

You do not have subscription access to this journal. Article tables are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access Optica Member Subscription

Equations (8)

You do not have subscription access to this journal. Equations are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access Optica Member Subscription

			Original Image Size
FFT Library	Compiler	Preferred Size	$256 \times 256$	$512 \times 512$	$1024 \times 1024$
FFTW	MSVC	No	13.41	14.55	176.44
		Yes	2.74	11.08	61.77
	ICC	No	9.75	10.48	153.35
MKL		Yes	2.94	10.33	56.73
	MSVC	No	12.20	13.63	174.36
		Yes	2.53	10.32	58.42
	ICC	No	9.17	10.38	153.31
		Yes	2.51	10.09	50.21

			Speedup	Speedup	Speedup
Item	Technique	Time (s)	Compared with A	Compared with B	Compared with C
A	MATLAB (sequential)	11916.59	–	–	–
B	MATLAB	3253.94	$3.66 \times$	–	–
C	C++ (sequential)	1549.89	$7.69 \times$	$2.10 \times$	–
D	MKL	665.41	$17.91 \times$	$4.89 \times$	$2.33 \times$
E	OpenMP	297.20	$40.10 \times$	$10.95 \times$	$5.21 \times$
F	Cilk Plus	318.35	$37.70 \times$	$10.29 \times$	$4.90 \times$
G	TBB	313.05	$37.43 \times$	$10.22 \times$	$4.87 \times$
H	PPL	316.13	$38.07 \times$	$10.39 \times$	$4.95 \times$

			Speedup	Speedup	Speedup
Item	Technique	Time (s)	Compared with A	Compared with B	Compared with C
A	MATLAB (sequential)	1136.08	–	–	–
B	MATLAB	703.70	$1.61 \times$	–	–
C	C++ (Sequential)	208.60	$5.45 \times$	$3.37 \times$	–
D	MKL	78.76	$14.42 \times$	$8.93 \times$	$2.65 \times$
E	OpenMP	40.04	$28.37 \times$	$17.57 \times$	$5.22 \times$
F	Cilk Plus	42.64	$26.91 \times$	$16.67 \times$	$4.94 \times$
G	TBB	42.75	$26.64 \times$	$16.50 \times$	$4.89 \times$
H	PPL	42.21	$26.57 \times$	$16.50 \times$	$4.87 \times$

	OpenMP		Cilk Plus		TBB		PPL
	WFR2	WFF2	WFR2	WFF2	WFR2	WFF2	WFR2	WFF2
LA	31	13	27	9	40	18	26	8
LD	0	0	0	0	2	2	2	2
LM	51	41	51	41	53	47	53	47
Total	82	54	78	50	85	67	81	57

			Original Image Size
FFT Library	Compiler	Preferred Size	$256 \times 256$	$512 \times 512$	$1024 \times 1024$
FFTW	MSVC	No	13.41	14.55	176.44
		Yes	2.74	11.08	61.77
	ICC	No	9.75	10.48	153.35
MKL		Yes	2.94	10.33	56.73
	MSVC	No	12.20	13.63	174.36
		Yes	2.53	10.32	58.42
	ICC	No	9.17	10.38	153.31
		Yes	2.51	10.09	50.21

Abstract

Cited By

Figures (2)

Tables (6)

Equations (8)

Applied Optics