GPGPU-based surface inspection from structured white light

Miguel Bordallo López, Karri Niemelä, Olli Silvén

Research output: Chapter in Book/Report/Conference proceedingConference article in proceedingsScientificpeer-review

1 Citation (Scopus)

Abstract

Automatic surface inspection has been used in the industry to reliably detect all kinds of surface defects and to measure the overall quality of a produced piece.
Structured light systems (SLS) are based on the reconstruction of the 3D information of a selected area by projecting several phase-shifted sinusoidal patterns onto a surface.
Due to the high speed of production lines, surface inspection systems require extremely fast imaging methods and lots of computational power.
The cost of such systems can easily become considerable. The use of standard PCs and Graphics Processing Units (GPUs) for data processing tasks facilitates the construction of cost-effective systems.
We present a parallel implementation of the required algorithms written in C with CUDA extensions. In our contribution, we describe the challenges of the design on a GPU, compared with a traditional CPU implementation.
We provide a qualitative evaluation of the results and a comparison of the algorithm speed performance on several platforms.
The system is able to compute two megapixels height maps with 100 micrometers spatial resolution in less than 200ms on a mid-budget laptop.
Our GPU implementation runs about ten times faster than our previous C code implementation.
Original languageEnglish
Title of host publicationProceedings of SPIE 8295
Subtitle of host publicationImage Processing, Algorithms and Systems X and Parallel Processing for Imaging Applications II
PublisherInternational Society for Optics and Photonics SPIE
ISBN (Print)978-0-8194-8942-5
DOIs
Publication statusPublished - 2012
MoE publication typeA4 Article in a conference publication
EventImage Processing: Algorithms and Systems X; and Parallel Processing for Imaging Applications II - Burlingame, California, United States
Duration: 22 Jan 201226 Jan 2012

Publication series

SeriesProceedings of SPIE
Volume8295
ISSN0277-786X

Conference

ConferenceImage Processing: Algorithms and Systems X; and Parallel Processing for Imaging Applications II
CountryUnited States
CityBurlingame, California
Period22/01/1226/01/12

Fingerprint

Inspection
Surface defects
Program processors
Costs
Imaging techniques
Graphics processing unit
Industry

Cite this

Bordallo López, M., Niemelä, K., & Silvén, O. (2012). GPGPU-based surface inspection from structured white light. In Proceedings of SPIE 8295: Image Processing, Algorithms and Systems X and Parallel Processing for Imaging Applications II [829510] International Society for Optics and Photonics SPIE. Proceedings of SPIE, Vol.. 8295 https://doi.org/10.1117/12.907349
Bordallo López, Miguel ; Niemelä, Karri ; Silvén, Olli. / GPGPU-based surface inspection from structured white light. Proceedings of SPIE 8295: Image Processing, Algorithms and Systems X and Parallel Processing for Imaging Applications II. International Society for Optics and Photonics SPIE, 2012. (Proceedings of SPIE, Vol. 8295).
@inproceedings{1ea4499d306e415fa7f1b3b7265b95d5,
title = "GPGPU-based surface inspection from structured white light",
abstract = "Automatic surface inspection has been used in the industry to reliably detect all kinds of surface defects and to measure the overall quality of a produced piece. Structured light systems (SLS) are based on the reconstruction of the 3D information of a selected area by projecting several phase-shifted sinusoidal patterns onto a surface. Due to the high speed of production lines, surface inspection systems require extremely fast imaging methods and lots of computational power. The cost of such systems can easily become considerable. The use of standard PCs and Graphics Processing Units (GPUs) for data processing tasks facilitates the construction of cost-effective systems. We present a parallel implementation of the required algorithms written in C with CUDA extensions. In our contribution, we describe the challenges of the design on a GPU, compared with a traditional CPU implementation. We provide a qualitative evaluation of the results and a comparison of the algorithm speed performance on several platforms. The system is able to compute two megapixels height maps with 100 micrometers spatial resolution in less than 200ms on a mid-budget laptop. Our GPU implementation runs about ten times faster than our previous C code implementation.",
author = "{Bordallo L{\'o}pez}, Miguel and Karri Niemel{\"a} and Olli Silv{\'e}n",
year = "2012",
doi = "10.1117/12.907349",
language = "English",
isbn = "978-0-8194-8942-5",
series = "Proceedings of SPIE",
publisher = "International Society for Optics and Photonics SPIE",
booktitle = "Proceedings of SPIE 8295",
address = "United States",

}

Bordallo López, M, Niemelä, K & Silvén, O 2012, GPGPU-based surface inspection from structured white light. in Proceedings of SPIE 8295: Image Processing, Algorithms and Systems X and Parallel Processing for Imaging Applications II., 829510, International Society for Optics and Photonics SPIE, Proceedings of SPIE, vol. 8295, Image Processing: Algorithms and Systems X; and Parallel Processing for Imaging Applications II, Burlingame, California, United States, 22/01/12. https://doi.org/10.1117/12.907349

GPGPU-based surface inspection from structured white light. / Bordallo López, Miguel; Niemelä, Karri; Silvén, Olli.

Proceedings of SPIE 8295: Image Processing, Algorithms and Systems X and Parallel Processing for Imaging Applications II. International Society for Optics and Photonics SPIE, 2012. 829510 (Proceedings of SPIE, Vol. 8295).

Research output: Chapter in Book/Report/Conference proceedingConference article in proceedingsScientificpeer-review

TY - GEN

T1 - GPGPU-based surface inspection from structured white light

AU - Bordallo López, Miguel

AU - Niemelä, Karri

AU - Silvén, Olli

PY - 2012

Y1 - 2012

N2 - Automatic surface inspection has been used in the industry to reliably detect all kinds of surface defects and to measure the overall quality of a produced piece. Structured light systems (SLS) are based on the reconstruction of the 3D information of a selected area by projecting several phase-shifted sinusoidal patterns onto a surface. Due to the high speed of production lines, surface inspection systems require extremely fast imaging methods and lots of computational power. The cost of such systems can easily become considerable. The use of standard PCs and Graphics Processing Units (GPUs) for data processing tasks facilitates the construction of cost-effective systems. We present a parallel implementation of the required algorithms written in C with CUDA extensions. In our contribution, we describe the challenges of the design on a GPU, compared with a traditional CPU implementation. We provide a qualitative evaluation of the results and a comparison of the algorithm speed performance on several platforms. The system is able to compute two megapixels height maps with 100 micrometers spatial resolution in less than 200ms on a mid-budget laptop. Our GPU implementation runs about ten times faster than our previous C code implementation.

AB - Automatic surface inspection has been used in the industry to reliably detect all kinds of surface defects and to measure the overall quality of a produced piece. Structured light systems (SLS) are based on the reconstruction of the 3D information of a selected area by projecting several phase-shifted sinusoidal patterns onto a surface. Due to the high speed of production lines, surface inspection systems require extremely fast imaging methods and lots of computational power. The cost of such systems can easily become considerable. The use of standard PCs and Graphics Processing Units (GPUs) for data processing tasks facilitates the construction of cost-effective systems. We present a parallel implementation of the required algorithms written in C with CUDA extensions. In our contribution, we describe the challenges of the design on a GPU, compared with a traditional CPU implementation. We provide a qualitative evaluation of the results and a comparison of the algorithm speed performance on several platforms. The system is able to compute two megapixels height maps with 100 micrometers spatial resolution in less than 200ms on a mid-budget laptop. Our GPU implementation runs about ten times faster than our previous C code implementation.

U2 - 10.1117/12.907349

DO - 10.1117/12.907349

M3 - Conference article in proceedings

SN - 978-0-8194-8942-5

T3 - Proceedings of SPIE

BT - Proceedings of SPIE 8295

PB - International Society for Optics and Photonics SPIE

ER -

Bordallo López M, Niemelä K, Silvén O. GPGPU-based surface inspection from structured white light. In Proceedings of SPIE 8295: Image Processing, Algorithms and Systems X and Parallel Processing for Imaging Applications II. International Society for Optics and Photonics SPIE. 2012. 829510. (Proceedings of SPIE, Vol. 8295). https://doi.org/10.1117/12.907349