IRNDT Inspection Via Sparse Principal Component Thermography

Bardia Yousefi, Stefano Sfarra, Fabrizio Sarasini, Xavier P.V. Maldague

Research output: Chapter in Book/Report/Conference proceedingConference contribution

3 Citations (Scopus)

Abstract

The recent applications in the field of thermography and Infrared Non-Destructive Testing (IRNDT) involved many different research fields, and in most of these applications well-known infrared approaches have been utilized for thermal image enhancement, thermal image segmentation, and particularly defect segmentation in IRNDT. Principal Component Analysis (PCA) or Principal Component Thermography (PCT) is one of these methods that has been countlessly used and it is unequivocally one of the constantly referred approaches in this field. Unfortunately, it suffers from being a linear transformation and besides that, finding appropriate basis through its eigenimage decomposition is the shortcoming. Here, an application of non-linear eigen decomposition using Sparse Principal Component Analysis/Thermography (Sparse-PCA or Sparse-PCT) is addressed for segmentation of defects inherent to two hybrid composites (carbon and flax fiber epoxy prepregs). The results indicate considerable segmentation performance when it is compared to similar approaches.

Original languageEnglish
Title of host publication2018 IEEE Canadian Conference on Electrical and Computer Engineering, CCECE 2018
PublisherInstitute of Electrical and Electronics Engineers Inc.
Volume2018-May
ISBN (Print)9781538624104
DOIs
Publication statusPublished - 27 Aug 2018
Event2018 IEEE Canadian Conference on Electrical and Computer Engineering, CCECE 2018 - Quebec City, Canada
Duration: 13 May 201816 May 2018

Conference

Conference2018 IEEE Canadian Conference on Electrical and Computer Engineering, CCECE 2018
CountryCanada
CityQuebec City
Period13.5.1816.5.18

Fingerprint

Nondestructive examination
Principal component analysis
Inspection
Infrared radiation
Decomposition
Flax
Defects
Linear transformations
Image enhancement
Image segmentation
Carbon
Fibers
Composite materials
Hot Temperature

Keywords

  • Image segmentation
  • Infrared non-destructive testing
  • Sparse principal component analy-sis/thermography

ASJC Scopus subject areas

  • Hardware and Architecture
  • Electrical and Electronic Engineering

Cite this

Yousefi, B., Sfarra, S., Sarasini, F., & Maldague, X. P. V. (2018). IRNDT Inspection Via Sparse Principal Component Thermography. In 2018 IEEE Canadian Conference on Electrical and Computer Engineering, CCECE 2018 (Vol. 2018-May). [8447887] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/CCECE.2018.8447887

IRNDT Inspection Via Sparse Principal Component Thermography. / Yousefi, Bardia; Sfarra, Stefano; Sarasini, Fabrizio; Maldague, Xavier P.V.

2018 IEEE Canadian Conference on Electrical and Computer Engineering, CCECE 2018. Vol. 2018-May Institute of Electrical and Electronics Engineers Inc., 2018. 8447887.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Yousefi, B, Sfarra, S, Sarasini, F & Maldague, XPV 2018, IRNDT Inspection Via Sparse Principal Component Thermography. in 2018 IEEE Canadian Conference on Electrical and Computer Engineering, CCECE 2018. vol. 2018-May, 8447887, Institute of Electrical and Electronics Engineers Inc., 2018 IEEE Canadian Conference on Electrical and Computer Engineering, CCECE 2018, Quebec City, Canada, 13.5.18. https://doi.org/10.1109/CCECE.2018.8447887
Yousefi B, Sfarra S, Sarasini F, Maldague XPV. IRNDT Inspection Via Sparse Principal Component Thermography. In 2018 IEEE Canadian Conference on Electrical and Computer Engineering, CCECE 2018. Vol. 2018-May. Institute of Electrical and Electronics Engineers Inc. 2018. 8447887 https://doi.org/10.1109/CCECE.2018.8447887
Yousefi, Bardia ; Sfarra, Stefano ; Sarasini, Fabrizio ; Maldague, Xavier P.V. / IRNDT Inspection Via Sparse Principal Component Thermography. 2018 IEEE Canadian Conference on Electrical and Computer Engineering, CCECE 2018. Vol. 2018-May Institute of Electrical and Electronics Engineers Inc., 2018.
@inproceedings{5852d60253444fab8324bc95cce52269,
title = "IRNDT Inspection Via Sparse Principal Component Thermography",
abstract = "The recent applications in the field of thermography and Infrared Non-Destructive Testing (IRNDT) involved many different research fields, and in most of these applications well-known infrared approaches have been utilized for thermal image enhancement, thermal image segmentation, and particularly defect segmentation in IRNDT. Principal Component Analysis (PCA) or Principal Component Thermography (PCT) is one of these methods that has been countlessly used and it is unequivocally one of the constantly referred approaches in this field. Unfortunately, it suffers from being a linear transformation and besides that, finding appropriate basis through its eigenimage decomposition is the shortcoming. Here, an application of non-linear eigen decomposition using Sparse Principal Component Analysis/Thermography (Sparse-PCA or Sparse-PCT) is addressed for segmentation of defects inherent to two hybrid composites (carbon and flax fiber epoxy prepregs). The results indicate considerable segmentation performance when it is compared to similar approaches.",
keywords = "Image segmentation, Infrared non-destructive testing, Sparse principal component analy-sis/thermography",
author = "Bardia Yousefi and Stefano Sfarra and Fabrizio Sarasini and Maldague, {Xavier P.V.}",
year = "2018",
month = "8",
day = "27",
doi = "10.1109/CCECE.2018.8447887",
language = "English",
isbn = "9781538624104",
volume = "2018-May",
booktitle = "2018 IEEE Canadian Conference on Electrical and Computer Engineering, CCECE 2018",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
address = "United States",

}

TY - GEN

T1 - IRNDT Inspection Via Sparse Principal Component Thermography

AU - Yousefi, Bardia

AU - Sfarra, Stefano

AU - Sarasini, Fabrizio

AU - Maldague, Xavier P.V.

PY - 2018/8/27

Y1 - 2018/8/27

N2 - The recent applications in the field of thermography and Infrared Non-Destructive Testing (IRNDT) involved many different research fields, and in most of these applications well-known infrared approaches have been utilized for thermal image enhancement, thermal image segmentation, and particularly defect segmentation in IRNDT. Principal Component Analysis (PCA) or Principal Component Thermography (PCT) is one of these methods that has been countlessly used and it is unequivocally one of the constantly referred approaches in this field. Unfortunately, it suffers from being a linear transformation and besides that, finding appropriate basis through its eigenimage decomposition is the shortcoming. Here, an application of non-linear eigen decomposition using Sparse Principal Component Analysis/Thermography (Sparse-PCA or Sparse-PCT) is addressed for segmentation of defects inherent to two hybrid composites (carbon and flax fiber epoxy prepregs). The results indicate considerable segmentation performance when it is compared to similar approaches.

AB - The recent applications in the field of thermography and Infrared Non-Destructive Testing (IRNDT) involved many different research fields, and in most of these applications well-known infrared approaches have been utilized for thermal image enhancement, thermal image segmentation, and particularly defect segmentation in IRNDT. Principal Component Analysis (PCA) or Principal Component Thermography (PCT) is one of these methods that has been countlessly used and it is unequivocally one of the constantly referred approaches in this field. Unfortunately, it suffers from being a linear transformation and besides that, finding appropriate basis through its eigenimage decomposition is the shortcoming. Here, an application of non-linear eigen decomposition using Sparse Principal Component Analysis/Thermography (Sparse-PCA or Sparse-PCT) is addressed for segmentation of defects inherent to two hybrid composites (carbon and flax fiber epoxy prepregs). The results indicate considerable segmentation performance when it is compared to similar approaches.

KW - Image segmentation

KW - Infrared non-destructive testing

KW - Sparse principal component analy-sis/thermography

UR - http://www.scopus.com/inward/record.url?scp=85053596446&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85053596446&partnerID=8YFLogxK

U2 - 10.1109/CCECE.2018.8447887

DO - 10.1109/CCECE.2018.8447887

M3 - Conference contribution

SN - 9781538624104

VL - 2018-May

BT - 2018 IEEE Canadian Conference on Electrical and Computer Engineering, CCECE 2018

PB - Institute of Electrical and Electronics Engineers Inc.

ER -