Decision-level fusion of spatially scattered multi-modal data for nondestructive inspection of surface defects

René Heideklang, Parisa Shokouhi

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

This article focuses on the fusion of flaw indications from multi-sensor nondestructive materials testing. Because each testing method makes use of a different physical principle, a multi-method approach has the potential of effectively differentiating actual defect indications from the many false alarms, thus enhancing detection reliability. In this study, we propose a new technique for aggregating scattered two- or three-dimensional sensory data. Using a density-based approach, the proposed method explicitly addresses localization uncertainties such as registration errors. This feature marks one of the major of advantages of this approach over pixel-based image fusion techniques. We provide guidelines on how to set all the key parameters and demonstrate the technique’s robustness. Finally, we apply our fusion approach to experimental data and demonstrate its capability to locate small defects by substantially reducing false alarms under conditions where no single-sensor method is adequate.

Original languageEnglish (US)
Article number105
JournalSensors (Switzerland)
Volume16
Issue number1
DOIs
StatePublished - Jan 15 2016

Fingerprint

Surface defects
surface defects
inspection
fusion
Inspection
false alarms
Defects
defects
indication
Materials testing
Image fusion
sensors
Sensors
Materials Testing
Pixels
pixels
Uncertainty
Testing
Guidelines

All Science Journal Classification (ASJC) codes

  • Analytical Chemistry
  • Atomic and Molecular Physics, and Optics
  • Biochemistry
  • Instrumentation
  • Electrical and Electronic Engineering

Cite this

@article{9e7c625335ec4f79901cbc8908d87bae,
title = "Decision-level fusion of spatially scattered multi-modal data for nondestructive inspection of surface defects",
abstract = "This article focuses on the fusion of flaw indications from multi-sensor nondestructive materials testing. Because each testing method makes use of a different physical principle, a multi-method approach has the potential of effectively differentiating actual defect indications from the many false alarms, thus enhancing detection reliability. In this study, we propose a new technique for aggregating scattered two- or three-dimensional sensory data. Using a density-based approach, the proposed method explicitly addresses localization uncertainties such as registration errors. This feature marks one of the major of advantages of this approach over pixel-based image fusion techniques. We provide guidelines on how to set all the key parameters and demonstrate the technique’s robustness. Finally, we apply our fusion approach to experimental data and demonstrate its capability to locate small defects by substantially reducing false alarms under conditions where no single-sensor method is adequate.",
author = "Ren{\'e} Heideklang and Parisa Shokouhi",
year = "2016",
month = "1",
day = "15",
doi = "10.3390/s16010105",
language = "English (US)",
volume = "16",
journal = "Sensors",
issn = "1424-3210",
publisher = "Multidisciplinary Digital Publishing Institute (MDPI)",
number = "1",

}

Decision-level fusion of spatially scattered multi-modal data for nondestructive inspection of surface defects. / Heideklang, René; Shokouhi, Parisa.

In: Sensors (Switzerland), Vol. 16, No. 1, 105, 15.01.2016.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Decision-level fusion of spatially scattered multi-modal data for nondestructive inspection of surface defects

AU - Heideklang, René

AU - Shokouhi, Parisa

PY - 2016/1/15

Y1 - 2016/1/15

N2 - This article focuses on the fusion of flaw indications from multi-sensor nondestructive materials testing. Because each testing method makes use of a different physical principle, a multi-method approach has the potential of effectively differentiating actual defect indications from the many false alarms, thus enhancing detection reliability. In this study, we propose a new technique for aggregating scattered two- or three-dimensional sensory data. Using a density-based approach, the proposed method explicitly addresses localization uncertainties such as registration errors. This feature marks one of the major of advantages of this approach over pixel-based image fusion techniques. We provide guidelines on how to set all the key parameters and demonstrate the technique’s robustness. Finally, we apply our fusion approach to experimental data and demonstrate its capability to locate small defects by substantially reducing false alarms under conditions where no single-sensor method is adequate.

AB - This article focuses on the fusion of flaw indications from multi-sensor nondestructive materials testing. Because each testing method makes use of a different physical principle, a multi-method approach has the potential of effectively differentiating actual defect indications from the many false alarms, thus enhancing detection reliability. In this study, we propose a new technique for aggregating scattered two- or three-dimensional sensory data. Using a density-based approach, the proposed method explicitly addresses localization uncertainties such as registration errors. This feature marks one of the major of advantages of this approach over pixel-based image fusion techniques. We provide guidelines on how to set all the key parameters and demonstrate the technique’s robustness. Finally, we apply our fusion approach to experimental data and demonstrate its capability to locate small defects by substantially reducing false alarms under conditions where no single-sensor method is adequate.

UR - http://www.scopus.com/inward/record.url?scp=84987677705&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84987677705&partnerID=8YFLogxK

U2 - 10.3390/s16010105

DO - 10.3390/s16010105

M3 - Article

C2 - 26784200

AN - SCOPUS:84987677705

VL - 16

JO - Sensors

JF - Sensors

SN - 1424-3210

IS - 1

M1 - 105

ER -