Orders-of-magnitude performance increases in GPU-accelerated correlation of images from the International Space Station

Peter J. Lu; Hidekazu Oki; Catherine A. Frey; Gregory E. Chamitoff; Leroy Chiao; Edward M. Fincke; C. Michael Foale; Sandra H. Magnus; William S. McArthur; Daniel M. Tani; Peggy A. Whitson; Jeffrey N. Williams; William V. Meyer; Ronald J. Sicker; Brion J. Au; Mark Christiansen; Andrew B. Schofield; David A. Weitz

doi:10.1007/s11554-009-0133-1

Orders-of-magnitude performance increases in GPU-accelerated correlation of images from the International Space Station

Peter J. Lu, Hidekazu Oki, Catherine A. Frey, Gregory E. Chamitoff, Leroy Chiao, Edward M. Fincke, C. Michael Foale, Sandra H. Magnus, William S. McArthur, Daniel M. Tani, Peggy A. Whitson, Jeffrey N. Williams, William V. Meyer, Ronald J. Sicker, Brion J. Au, Mark Christiansen, Andrew B. Schofield, David A. Weitz

School of Physics and Astronomy

Research output: Contribution to journal › Article › peer-review

Abstract

We implement image correlation, a fundamental component of many real-time imaging and tracking systems, on a graphics processing unit (GPU) using NVI-DIA's CUDA platform. We use our code to analyze images of liquid-gas phase separation in a model colloid-polymer system, photographed in the absence of gravity aboard the International Space Station (ISS). Our GPU code is 4,000 times faster than simple MATLAB code performing the same calculation on a central processing unit (CPU), 130 times faster than simple C code, and 30 times faster than optimized C++ code using single-instruction, multiple-data (SIMD) extensions. The speed increases from these parallel algorithms enable us to analyze images downlinked from the ISS in a rapid fashion and send feedback to astronauts on orbit while the experiments are still being run.

Original language	English
Pages (from-to)	179-193
Number of pages	15
Journal	Journal of real-Time image processing
Volume	5
Issue number	3
DOIs	https://doi.org/10.1007/s11554-009-0133-1
Publication status	Published - Sept 2010

Keywords / Materials (for Non-textual outputs)

GPU
CUDA
Autocorrelation
International Space Station
SIMD
GRAPHICS PROCESSING UNITS
COLLOID POLYMER MIXTURE
SIMULATIONS
MICROSCOPY

Access to Document

10.1007/s11554-009-0133-1

Cite this

Lu, P. J., Oki, H., Frey, C. A., Chamitoff, G. E., Chiao, L., Fincke, E. M., Foale, C. M., Magnus, S. H., McArthur, W. S., Tani, D. M., Whitson, P. A., Williams, J. N., Meyer, W. V., Sicker, R. J., Au, B. J., Christiansen, M., Schofield, A. B., & Weitz, D. A. (2010). Orders-of-magnitude performance increases in GPU-accelerated correlation of images from the International Space Station. Journal of real-Time image processing, 5(3), 179-193. https://doi.org/10.1007/s11554-009-0133-1

@article{5b6e98427a65433b8e23877377586a60,

title = "Orders-of-magnitude performance increases in GPU-accelerated correlation of images from the International Space Station",

abstract = "We implement image correlation, a fundamental component of many real-time imaging and tracking systems, on a graphics processing unit (GPU) using NVI-DIA's CUDA platform. We use our code to analyze images of liquid-gas phase separation in a model colloid-polymer system, photographed in the absence of gravity aboard the International Space Station (ISS). Our GPU code is 4,000 times faster than simple MATLAB code performing the same calculation on a central processing unit (CPU), 130 times faster than simple C code, and 30 times faster than optimized C++ code using single-instruction, multiple-data (SIMD) extensions. The speed increases from these parallel algorithms enable us to analyze images downlinked from the ISS in a rapid fashion and send feedback to astronauts on orbit while the experiments are still being run.",

keywords = "GPU, CUDA, Autocorrelation, International Space Station, SIMD, GRAPHICS PROCESSING UNITS, COLLOID POLYMER MIXTURE, SIMULATIONS, MICROSCOPY",

author = "Lu, \{Peter J.\} and Hidekazu Oki and Frey, \{Catherine A.\} and Chamitoff, \{Gregory E.\} and Leroy Chiao and Fincke, \{Edward M.\} and Foale, \{C. Michael\} and Magnus, \{Sandra H.\} and McArthur, \{William S.\} and Tani, \{Daniel M.\} and Whitson, \{Peggy A.\} and Williams, \{Jeffrey N.\} and Meyer, \{William V.\} and Sicker, \{Ronald J.\} and Au, \{Brion J.\} and Mark Christiansen and Schofield, \{Andrew B.\} and Weitz, \{David A.\}",

year = "2010",

month = sep,

doi = "10.1007/s11554-009-0133-1",

language = "English",

volume = "5",

pages = "179--193",

journal = "Journal of real-Time image processing",

issn = "1861-8200",

publisher = "Springer Nature",

number = "3",

}

Lu, PJ, Oki, H, Frey, CA, Chamitoff, GE, Chiao, L, Fincke, EM, Foale, CM, Magnus, SH, McArthur, WS, Tani, DM, Whitson, PA, Williams, JN, Meyer, WV, Sicker, RJ, Au, BJ, Christiansen, M, Schofield, AB & Weitz, DA 2010, 'Orders-of-magnitude performance increases in GPU-accelerated correlation of images from the International Space Station', Journal of real-Time image processing, vol. 5, no. 3, pp. 179-193. https://doi.org/10.1007/s11554-009-0133-1

TY - JOUR

T1 - Orders-of-magnitude performance increases in GPU-accelerated correlation of images from the International Space Station

AU - Lu, Peter J.

AU - Oki, Hidekazu

AU - Frey, Catherine A.

AU - Chamitoff, Gregory E.

AU - Chiao, Leroy

AU - Fincke, Edward M.

AU - Foale, C. Michael

AU - Magnus, Sandra H.

AU - McArthur, William S.

AU - Tani, Daniel M.

AU - Whitson, Peggy A.

AU - Williams, Jeffrey N.

AU - Meyer, William V.

AU - Sicker, Ronald J.

AU - Au, Brion J.

AU - Christiansen, Mark

AU - Schofield, Andrew B.

AU - Weitz, David A.

PY - 2010/9

Y1 - 2010/9

N2 - We implement image correlation, a fundamental component of many real-time imaging and tracking systems, on a graphics processing unit (GPU) using NVI-DIA's CUDA platform. We use our code to analyze images of liquid-gas phase separation in a model colloid-polymer system, photographed in the absence of gravity aboard the International Space Station (ISS). Our GPU code is 4,000 times faster than simple MATLAB code performing the same calculation on a central processing unit (CPU), 130 times faster than simple C code, and 30 times faster than optimized C++ code using single-instruction, multiple-data (SIMD) extensions. The speed increases from these parallel algorithms enable us to analyze images downlinked from the ISS in a rapid fashion and send feedback to astronauts on orbit while the experiments are still being run.

AB - We implement image correlation, a fundamental component of many real-time imaging and tracking systems, on a graphics processing unit (GPU) using NVI-DIA's CUDA platform. We use our code to analyze images of liquid-gas phase separation in a model colloid-polymer system, photographed in the absence of gravity aboard the International Space Station (ISS). Our GPU code is 4,000 times faster than simple MATLAB code performing the same calculation on a central processing unit (CPU), 130 times faster than simple C code, and 30 times faster than optimized C++ code using single-instruction, multiple-data (SIMD) extensions. The speed increases from these parallel algorithms enable us to analyze images downlinked from the ISS in a rapid fashion and send feedback to astronauts on orbit while the experiments are still being run.

KW - GPU

KW - CUDA

KW - Autocorrelation

KW - International Space Station

KW - SIMD

KW - GRAPHICS PROCESSING UNITS

KW - COLLOID POLYMER MIXTURE

KW - SIMULATIONS

KW - MICROSCOPY

U2 - 10.1007/s11554-009-0133-1

DO - 10.1007/s11554-009-0133-1

M3 - Article

SN - 1861-8200

VL - 5

SP - 179

EP - 193

JO - Journal of real-Time image processing

JF - Journal of real-Time image processing

IS - 3

ER -

Orders-of-magnitude performance increases in GPU-accelerated correlation of images from the International Space Station

Abstract

Keywords / Materials (for Non-textual outputs)

Access to Document

Fingerprint

Cite this