Parallel connected-component labeling algorithm for GPGPU applications

In Yong Jung; Chang Sung Jeong

doi:10.1109/ISCIT.2010.5665161

Parallel connected-component labeling algorithm for GPGPU applications

In Yong Jung, Chang Sung Jeong

School of Electrical Engineering

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

13 Citations (Scopus)

Abstract

This paper proposes a new connected component labeling algorithm for GPGPU applications based on NVIDIA's CUDA. Various approaches and algorithms for connected component labeling with minimal execution time were designed, but the most of them have been focused on optimizing CPU algorithm. Therefore it is hard to apply these approaches to GPGPU programming models such as NVIDIA's CUDA. Today, GPGPU (General Purpose Graphic Processing Unit) technologies offer dedicated parallel hardware and programming model, and many applications are being moved onto the GPGPU. This algorithm is a multi-pass algorithm to utilize for GPGPU applications, and evaluation results show that maximum speedup is more than double compared with conventional CPU algorithms.

Original language	English
Title of host publication	ISCIT 2010 - 2010 10th International Symposium on Communications and Information Technologies
Pages	1149-1153
Number of pages	5
DOIs	https://doi.org/10.1109/ISCIT.2010.5665161
Publication status	Published - 2010
Event	2010 10th International Symposium on Communications and Information Technologies, ISCIT 2010 - Tokyo, Japan Duration: 2010 Oct 26 → 2010 Oct 29

Publication series

Name	ISCIT 2010 - 2010 10th International Symposium on Communications and Information Technologies

Other

Other	2010 10th International Symposium on Communications and Information Technologies, ISCIT 2010
Country/Territory	Japan
City	Tokyo
Period	10/10/26 → 10/10/29

ASJC Scopus subject areas

Computer Networks and Communications
Information Systems

Access to Document

10.1109/ISCIT.2010.5665161

Cite this

Jung, I. Y., & Jeong, C. S. (2010). Parallel connected-component labeling algorithm for GPGPU applications. In ISCIT 2010 - 2010 10th International Symposium on Communications and Information Technologies (pp. 1149-1153). [5665161] (ISCIT 2010 - 2010 10th International Symposium on Communications and Information Technologies). https://doi.org/10.1109/ISCIT.2010.5665161

Parallel connected-component labeling algorithm for GPGPU applications. / Jung, In Yong; Jeong, Chang Sung.
ISCIT 2010 - 2010 10th International Symposium on Communications and Information Technologies. 2010. p. 1149-1153 5665161 (ISCIT 2010 - 2010 10th International Symposium on Communications and Information Technologies).

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

Jung, IY & Jeong, CS 2010, Parallel connected-component labeling algorithm for GPGPU applications. in ISCIT 2010 - 2010 10th International Symposium on Communications and Information Technologies., 5665161, ISCIT 2010 - 2010 10th International Symposium on Communications and Information Technologies, pp. 1149-1153, 2010 10th International Symposium on Communications and Information Technologies, ISCIT 2010, Tokyo, Japan, 10/10/26. https://doi.org/10.1109/ISCIT.2010.5665161

@inproceedings{a7548311ce914c1ab15891a684279bad,

title = "Parallel connected-component labeling algorithm for GPGPU applications",

abstract = "This paper proposes a new connected component labeling algorithm for GPGPU applications based on NVIDIA's CUDA. Various approaches and algorithms for connected component labeling with minimal execution time were designed, but the most of them have been focused on optimizing CPU algorithm. Therefore it is hard to apply these approaches to GPGPU programming models such as NVIDIA's CUDA. Today, GPGPU (General Purpose Graphic Processing Unit) technologies offer dedicated parallel hardware and programming model, and many applications are being moved onto the GPGPU. This algorithm is a multi-pass algorithm to utilize for GPGPU applications, and evaluation results show that maximum speedup is more than double compared with conventional CPU algorithms.",

author = "Jung, {In Yong} and Jeong, {Chang Sung}",

year = "2010",

doi = "10.1109/ISCIT.2010.5665161",

language = "English",

isbn = "9781424470105",

series = "ISCIT 2010 - 2010 10th International Symposium on Communications and Information Technologies",

pages = "1149--1153",

booktitle = "ISCIT 2010 - 2010 10th International Symposium on Communications and Information Technologies",

note = "2010 10th International Symposium on Communications and Information Technologies, ISCIT 2010 ; Conference date: 26-10-2010 Through 29-10-2010",

}

TY - GEN

T1 - Parallel connected-component labeling algorithm for GPGPU applications

AU - Jung, In Yong

AU - Jeong, Chang Sung

PY - 2010

Y1 - 2010

N2 - This paper proposes a new connected component labeling algorithm for GPGPU applications based on NVIDIA's CUDA. Various approaches and algorithms for connected component labeling with minimal execution time were designed, but the most of them have been focused on optimizing CPU algorithm. Therefore it is hard to apply these approaches to GPGPU programming models such as NVIDIA's CUDA. Today, GPGPU (General Purpose Graphic Processing Unit) technologies offer dedicated parallel hardware and programming model, and many applications are being moved onto the GPGPU. This algorithm is a multi-pass algorithm to utilize for GPGPU applications, and evaluation results show that maximum speedup is more than double compared with conventional CPU algorithms.

AB - This paper proposes a new connected component labeling algorithm for GPGPU applications based on NVIDIA's CUDA. Various approaches and algorithms for connected component labeling with minimal execution time were designed, but the most of them have been focused on optimizing CPU algorithm. Therefore it is hard to apply these approaches to GPGPU programming models such as NVIDIA's CUDA. Today, GPGPU (General Purpose Graphic Processing Unit) technologies offer dedicated parallel hardware and programming model, and many applications are being moved onto the GPGPU. This algorithm is a multi-pass algorithm to utilize for GPGPU applications, and evaluation results show that maximum speedup is more than double compared with conventional CPU algorithms.

UR - http://www.scopus.com/inward/record.url?scp=78651242951&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=78651242951&partnerID=8YFLogxK

U2 - 10.1109/ISCIT.2010.5665161

DO - 10.1109/ISCIT.2010.5665161

M3 - Conference contribution

AN - SCOPUS:78651242951

SN - 9781424470105

T3 - ISCIT 2010 - 2010 10th International Symposium on Communications and Information Technologies

SP - 1149

EP - 1153

BT - ISCIT 2010 - 2010 10th International Symposium on Communications and Information Technologies

T2 - 2010 10th International Symposium on Communications and Information Technologies, ISCIT 2010

Y2 - 26 October 2010 through 29 October 2010

ER -

Parallel connected-component labeling algorithm for GPGPU applications

Abstract

Publication series

Other

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this