Accelerated catadioptric omnidirectional view image unwrapping processing using GPU parallelisation

Nguan Soon Chong, M. L Dennis Wong, Yau Hee Kho

Research output: Contribution to journalArticle

Abstract

Catadioptric omnidirectional view sensors have found increasing adoption in various robotic and surveillance applications due to their 360° field of view. However, the inherent distortion caused by the sensors prevents their direct utilisations using existing image processing techniques developed for perspective images. Therefore, a correction processing known as "unwrapping" is commonly performed. However, the unwrapping process incurs additional computational loads on central processing units. In this paper, a method to reduce this burden in the computation is investigated by exploiting the parallelism of graphical processing units (GPUs) based on the Compute Unified Device Architecture (CUDA). More specifically, we first introduce a general approach of parallelisation to the said process. Then, a series of adaptations to the CUDA platform is proposed to enable an optimised usage of the hardware platform. Finally, the performances of the unwrapping function were evaluated on a high-end and low-end GPU to demonstrate the effectiveness of the parallelisation approach.

Original languageEnglish
Pages (from-to)1-15
Number of pages15
JournalJournal of Real-Time Image Processing
DOIs
Publication statusAccepted/In press - 2013

Fingerprint

Image processing
Processing
Sensors
Program processors
Robotics
Hardware

Keywords

  • Bilinear interpolation
  • CUDA
  • GPU
  • Image unwrapping
  • Omnidirectional sensor
  • Parallelisation

ASJC Scopus subject areas

  • Information Systems

Cite this

Accelerated catadioptric omnidirectional view image unwrapping processing using GPU parallelisation. / Chong, Nguan Soon; Wong, M. L Dennis; Kho, Yau Hee.

In: Journal of Real-Time Image Processing, 2013, p. 1-15.

Research output: Contribution to journalArticle

@article{ee533a700f404efc9eb7205b94e5150f,
title = "Accelerated catadioptric omnidirectional view image unwrapping processing using GPU parallelisation",
abstract = "Catadioptric omnidirectional view sensors have found increasing adoption in various robotic and surveillance applications due to their 360° field of view. However, the inherent distortion caused by the sensors prevents their direct utilisations using existing image processing techniques developed for perspective images. Therefore, a correction processing known as {"}unwrapping{"} is commonly performed. However, the unwrapping process incurs additional computational loads on central processing units. In this paper, a method to reduce this burden in the computation is investigated by exploiting the parallelism of graphical processing units (GPUs) based on the Compute Unified Device Architecture (CUDA). More specifically, we first introduce a general approach of parallelisation to the said process. Then, a series of adaptations to the CUDA platform is proposed to enable an optimised usage of the hardware platform. Finally, the performances of the unwrapping function were evaluated on a high-end and low-end GPU to demonstrate the effectiveness of the parallelisation approach.",
keywords = "Bilinear interpolation, CUDA, GPU, Image unwrapping, Omnidirectional sensor, Parallelisation",
author = "Chong, {Nguan Soon} and Wong, {M. L Dennis} and Kho, {Yau Hee}",
year = "2013",
doi = "10.1007/s11554-013-0390-x",
language = "English",
pages = "1--15",
journal = "Journal of Real-Time Image Processing",
issn = "1861-8200",
publisher = "Springer Verlag",

}

TY - JOUR

T1 - Accelerated catadioptric omnidirectional view image unwrapping processing using GPU parallelisation

AU - Chong, Nguan Soon

AU - Wong, M. L Dennis

AU - Kho, Yau Hee

PY - 2013

Y1 - 2013

N2 - Catadioptric omnidirectional view sensors have found increasing adoption in various robotic and surveillance applications due to their 360° field of view. However, the inherent distortion caused by the sensors prevents their direct utilisations using existing image processing techniques developed for perspective images. Therefore, a correction processing known as "unwrapping" is commonly performed. However, the unwrapping process incurs additional computational loads on central processing units. In this paper, a method to reduce this burden in the computation is investigated by exploiting the parallelism of graphical processing units (GPUs) based on the Compute Unified Device Architecture (CUDA). More specifically, we first introduce a general approach of parallelisation to the said process. Then, a series of adaptations to the CUDA platform is proposed to enable an optimised usage of the hardware platform. Finally, the performances of the unwrapping function were evaluated on a high-end and low-end GPU to demonstrate the effectiveness of the parallelisation approach.

AB - Catadioptric omnidirectional view sensors have found increasing adoption in various robotic and surveillance applications due to their 360° field of view. However, the inherent distortion caused by the sensors prevents their direct utilisations using existing image processing techniques developed for perspective images. Therefore, a correction processing known as "unwrapping" is commonly performed. However, the unwrapping process incurs additional computational loads on central processing units. In this paper, a method to reduce this burden in the computation is investigated by exploiting the parallelism of graphical processing units (GPUs) based on the Compute Unified Device Architecture (CUDA). More specifically, we first introduce a general approach of parallelisation to the said process. Then, a series of adaptations to the CUDA platform is proposed to enable an optimised usage of the hardware platform. Finally, the performances of the unwrapping function were evaluated on a high-end and low-end GPU to demonstrate the effectiveness of the parallelisation approach.

KW - Bilinear interpolation

KW - CUDA

KW - GPU

KW - Image unwrapping

KW - Omnidirectional sensor

KW - Parallelisation

UR - http://www.scopus.com/inward/record.url?scp=84890893138&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84890893138&partnerID=8YFLogxK

U2 - 10.1007/s11554-013-0390-x

DO - 10.1007/s11554-013-0390-x

M3 - Article

SP - 1

EP - 15

JO - Journal of Real-Time Image Processing

JF - Journal of Real-Time Image Processing

SN - 1861-8200

ER -