GPU-Accelerated nick local image thresholding algorithm

M. Hassan Najafi; Anirudh Murali; David J. Lilja; John Sartori

doi:10.1109/ICPADS.2015.78

GPU-Accelerated nick local image thresholding algorithm

M. Hassan Najafi, Anirudh Murali, David J. Lilja, John Sartori

Electrical and Computer Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

4 Scopus citations

Abstract

Binarization plays an important role in document image processing, particularly in degraded document images. Among all local adaptive image thresholding algorithms, the Nick method has shown excellent binarization performance for degraded document images. However, local image thresholding algorithms, including the Nick method, are computationally intensive, requiring significant time to process input images. In this paper, we propose three CUDA GPU parallel implementations of the Nick local image thresholding algorithm for faster binarization of large images. Our experimental results show that the GPU-accelerated implementations of the Nick method can achieve up to 150x performance speedup on a GeForce GTX 480 compared to its optimized sequential implementation.

Original language	English (US)
Title of host publication	Proceedings - 2015 IEEE 21st International Conference on Parallel and Distributed Systems, ICPADS 2015
Publisher	IEEE Computer Society
Pages	576-584
Number of pages	9
ISBN (Electronic)	9780769557854
DOIs	https://doi.org/10.1109/ICPADS.2015.78
State	Published - Jan 15 2016
Event	21st IEEE International Conference on Parallel and Distributed Systems, ICPADS 2015 - Melbourne, Australia Duration: Dec 14 2015 → Dec 17 2015

Publication series

Name	Proceedings of the International Conference on Parallel and Distributed Systems - ICPADS
Volume	2016-January
ISSN (Print)	1521-9097

Other

Other	21st IEEE International Conference on Parallel and Distributed Systems, ICPADS 2015
Country/Territory	Australia
City	Melbourne
Period	12/14/15 → 12/17/15

Bibliographical note

Publisher Copyright:
© 2015 IEEE.

Keywords

CUDA GPU Programming
GPU acceleration
Image binarization
Image thresholding
Parallel programming

Access

10.1109/ICPADS.2015.78

OpenUrl availability

Full text

Cite this

Najafi, M. H., Murali, A., Lilja, D. J., & Sartori, J. (2016). GPU-Accelerated nick local image thresholding algorithm. In Proceedings - 2015 IEEE 21st International Conference on Parallel and Distributed Systems, ICPADS 2015 (pp. 576-584). Article 7384341 (Proceedings of the International Conference on Parallel and Distributed Systems - ICPADS; Vol. 2016-January). IEEE Computer Society. https://doi.org/10.1109/ICPADS.2015.78

GPU-Accelerated nick local image thresholding algorithm. / Najafi, M. Hassan; Murali, Anirudh; Lilja, David J. et al.
Proceedings - 2015 IEEE 21st International Conference on Parallel and Distributed Systems, ICPADS 2015. IEEE Computer Society, 2016. p. 576-584 7384341 (Proceedings of the International Conference on Parallel and Distributed Systems - ICPADS; Vol. 2016-January).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Najafi, MH, Murali, A, Lilja, DJ & Sartori, J 2016, GPU-Accelerated nick local image thresholding algorithm. in Proceedings - 2015 IEEE 21st International Conference on Parallel and Distributed Systems, ICPADS 2015., 7384341, Proceedings of the International Conference on Parallel and Distributed Systems - ICPADS, vol. 2016-January, IEEE Computer Society, pp. 576-584, 21st IEEE International Conference on Parallel and Distributed Systems, ICPADS 2015, Melbourne, Australia, 12/14/15. https://doi.org/10.1109/ICPADS.2015.78

Najafi MH, Murali A, Lilja DJ, Sartori J. GPU-Accelerated nick local image thresholding algorithm. In Proceedings - 2015 IEEE 21st International Conference on Parallel and Distributed Systems, ICPADS 2015. IEEE Computer Society. 2016. p. 576-584. 7384341. (Proceedings of the International Conference on Parallel and Distributed Systems - ICPADS). doi: 10.1109/ICPADS.2015.78

@inproceedings{af55f74b107f4aebb222d9b79be2fdd7,

title = "GPU-Accelerated nick local image thresholding algorithm",

abstract = "Binarization plays an important role in document image processing, particularly in degraded document images. Among all local adaptive image thresholding algorithms, the Nick method has shown excellent binarization performance for degraded document images. However, local image thresholding algorithms, including the Nick method, are computationally intensive, requiring significant time to process input images. In this paper, we propose three CUDA GPU parallel implementations of the Nick local image thresholding algorithm for faster binarization of large images. Our experimental results show that the GPU-accelerated implementations of the Nick method can achieve up to 150x performance speedup on a GeForce GTX 480 compared to its optimized sequential implementation.",

keywords = "CUDA GPU Programming, GPU acceleration, Image binarization, Image thresholding, Parallel programming",

author = "Najafi, {M. Hassan} and Anirudh Murali and Lilja, {David J.} and John Sartori",

note = "Publisher Copyright: {\textcopyright} 2015 IEEE.; 21st IEEE International Conference on Parallel and Distributed Systems, ICPADS 2015 ; Conference date: 14-12-2015 Through 17-12-2015",

year = "2016",

month = jan,

day = "15",

doi = "10.1109/ICPADS.2015.78",

language = "English (US)",

series = "Proceedings of the International Conference on Parallel and Distributed Systems - ICPADS",

publisher = "IEEE Computer Society",

pages = "576--584",

booktitle = "Proceedings - 2015 IEEE 21st International Conference on Parallel and Distributed Systems, ICPADS 2015",

}

TY - GEN

T1 - GPU-Accelerated nick local image thresholding algorithm

AU - Najafi, M. Hassan

AU - Murali, Anirudh

AU - Lilja, David J.

AU - Sartori, John

PY - 2016/1/15

Y1 - 2016/1/15

N2 - Binarization plays an important role in document image processing, particularly in degraded document images. Among all local adaptive image thresholding algorithms, the Nick method has shown excellent binarization performance for degraded document images. However, local image thresholding algorithms, including the Nick method, are computationally intensive, requiring significant time to process input images. In this paper, we propose three CUDA GPU parallel implementations of the Nick local image thresholding algorithm for faster binarization of large images. Our experimental results show that the GPU-accelerated implementations of the Nick method can achieve up to 150x performance speedup on a GeForce GTX 480 compared to its optimized sequential implementation.

AB - Binarization plays an important role in document image processing, particularly in degraded document images. Among all local adaptive image thresholding algorithms, the Nick method has shown excellent binarization performance for degraded document images. However, local image thresholding algorithms, including the Nick method, are computationally intensive, requiring significant time to process input images. In this paper, we propose three CUDA GPU parallel implementations of the Nick local image thresholding algorithm for faster binarization of large images. Our experimental results show that the GPU-accelerated implementations of the Nick method can achieve up to 150x performance speedup on a GeForce GTX 480 compared to its optimized sequential implementation.

KW - CUDA GPU Programming

KW - GPU acceleration

KW - Image binarization

KW - Image thresholding

KW - Parallel programming

UR - http://www.scopus.com/inward/record.url?scp=84964684010&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84964684010&partnerID=8YFLogxK

U2 - 10.1109/ICPADS.2015.78

DO - 10.1109/ICPADS.2015.78

M3 - Conference contribution

AN - SCOPUS:84964684010

T3 - Proceedings of the International Conference on Parallel and Distributed Systems - ICPADS

SP - 576

EP - 584

BT - Proceedings - 2015 IEEE 21st International Conference on Parallel and Distributed Systems, ICPADS 2015

PB - IEEE Computer Society

T2 - 21st IEEE International Conference on Parallel and Distributed Systems, ICPADS 2015

Y2 - 14 December 2015 through 17 December 2015

ER -

GPU-Accelerated nick local image thresholding algorithm

Abstract

Publication series

Other

Bibliographical note

Keywords

Access

OpenUrl availability

Other files and links

Fingerprint

Cite this