Importance of spatial and spectral data reduction in the detection of internal defects in food products

Xuechen Zhang, Christian Nansen, Nader Aryamanesh, Guijun Yan, Farid Boussaid

Research output: Contribution to journalArticle

12 Citations (Scopus)

Abstract

Despite the importance of data reduction as part of the processing of reflection-based classifications, this study represents one of the first in which the effects of both spatial and spectral data reductions on classification accuracies are quantified. Furthermore, the effects of approaches to data reduction were quantified for two separate classification methods, linear discriminant analysis (LDA) and support vector machine (SVM). As the model dataset, reflection data were acquired using a hyperspectral camera in 230 spectral channels from 401 to 879 nm (spectral resolution of 2.1 nm) from field pea (Pisum sativum) samples with and without internal pea weevil (Bruchus pisorum) infestation. We deployed five levels of spatial data reduction (binning) and eight levels of spectral data reduction (40 datasets). Forward stepwise LDA was used to select and include only spectral channels contributing the most to the separation of pixels from non-infested and infested field peas. Classification accuracies obtained with LDA and SVM were based on the classification of independent validation datasets. Overall, SVMs had significantly higher classification accuracies than LDAs (P < 0.01). There was a negative association between pixel resolution and classification accuracy, while spectral binning equivalent to up to 98% data reduction had negligible effect on classification accuracies. This study supports the potential use of reflection-based technologies in the quality control of food products with internal defects, and it highlights that spatial and spectral data reductions can (1) improve classification accuracies, (2) vastly decrease computer constraints, and (3) reduce analytical concerns associated with classifications of large and high-dimensional datasets.

LanguageEnglish
Pages473-480
Number of pages8
JournalApplied Spectroscopy
Volume69
Issue number4
DOIs
Publication statusPublished - 1 Jan 2015
Externally publishedYes

Keywords

  • Field pea
  • Hyperspectral imaging
  • Machine vision
  • Pea weevil
  • Reflection data
  • Spatial resolution
  • Spectral resolution

ASJC Scopus subject areas

  • Instrumentation
  • Spectroscopy

Cite this

Zhang, Xuechen ; Nansen, Christian ; Aryamanesh, Nader ; Yan, Guijun ; Boussaid, Farid. / Importance of spatial and spectral data reduction in the detection of internal defects in food products. In: Applied Spectroscopy. 2015 ; Vol. 69, No. 4. pp. 473-480.
@article{87859c534fe745a6a62025ff26198341,
title = "Importance of spatial and spectral data reduction in the detection of internal defects in food products",
abstract = "Despite the importance of data reduction as part of the processing of reflection-based classifications, this study represents one of the first in which the effects of both spatial and spectral data reductions on classification accuracies are quantified. Furthermore, the effects of approaches to data reduction were quantified for two separate classification methods, linear discriminant analysis (LDA) and support vector machine (SVM). As the model dataset, reflection data were acquired using a hyperspectral camera in 230 spectral channels from 401 to 879 nm (spectral resolution of 2.1 nm) from field pea (Pisum sativum) samples with and without internal pea weevil (Bruchus pisorum) infestation. We deployed five levels of spatial data reduction (binning) and eight levels of spectral data reduction (40 datasets). Forward stepwise LDA was used to select and include only spectral channels contributing the most to the separation of pixels from non-infested and infested field peas. Classification accuracies obtained with LDA and SVM were based on the classification of independent validation datasets. Overall, SVMs had significantly higher classification accuracies than LDAs (P < 0.01). There was a negative association between pixel resolution and classification accuracy, while spectral binning equivalent to up to 98{\%} data reduction had negligible effect on classification accuracies. This study supports the potential use of reflection-based technologies in the quality control of food products with internal defects, and it highlights that spatial and spectral data reductions can (1) improve classification accuracies, (2) vastly decrease computer constraints, and (3) reduce analytical concerns associated with classifications of large and high-dimensional datasets.",
keywords = "Field pea, Hyperspectral imaging, Machine vision, Pea weevil, Reflection data, Spatial resolution, Spectral resolution",
author = "Xuechen Zhang and Christian Nansen and Nader Aryamanesh and Guijun Yan and Farid Boussaid",
year = "2015",
month = "1",
day = "1",
doi = "10.1366/14-07672",
language = "English",
volume = "69",
pages = "473--480",
journal = "Applied Spectroscopy",
issn = "0003-7028",
publisher = "SAGE Publications Inc.",
number = "4",

}

Importance of spatial and spectral data reduction in the detection of internal defects in food products. / Zhang, Xuechen; Nansen, Christian; Aryamanesh, Nader; Yan, Guijun; Boussaid, Farid.

In: Applied Spectroscopy, Vol. 69, No. 4, 01.01.2015, p. 473-480.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Importance of spatial and spectral data reduction in the detection of internal defects in food products

AU - Zhang, Xuechen

AU - Nansen, Christian

AU - Aryamanesh, Nader

AU - Yan, Guijun

AU - Boussaid, Farid

PY - 2015/1/1

Y1 - 2015/1/1

N2 - Despite the importance of data reduction as part of the processing of reflection-based classifications, this study represents one of the first in which the effects of both spatial and spectral data reductions on classification accuracies are quantified. Furthermore, the effects of approaches to data reduction were quantified for two separate classification methods, linear discriminant analysis (LDA) and support vector machine (SVM). As the model dataset, reflection data were acquired using a hyperspectral camera in 230 spectral channels from 401 to 879 nm (spectral resolution of 2.1 nm) from field pea (Pisum sativum) samples with and without internal pea weevil (Bruchus pisorum) infestation. We deployed five levels of spatial data reduction (binning) and eight levels of spectral data reduction (40 datasets). Forward stepwise LDA was used to select and include only spectral channels contributing the most to the separation of pixels from non-infested and infested field peas. Classification accuracies obtained with LDA and SVM were based on the classification of independent validation datasets. Overall, SVMs had significantly higher classification accuracies than LDAs (P < 0.01). There was a negative association between pixel resolution and classification accuracy, while spectral binning equivalent to up to 98% data reduction had negligible effect on classification accuracies. This study supports the potential use of reflection-based technologies in the quality control of food products with internal defects, and it highlights that spatial and spectral data reductions can (1) improve classification accuracies, (2) vastly decrease computer constraints, and (3) reduce analytical concerns associated with classifications of large and high-dimensional datasets.

AB - Despite the importance of data reduction as part of the processing of reflection-based classifications, this study represents one of the first in which the effects of both spatial and spectral data reductions on classification accuracies are quantified. Furthermore, the effects of approaches to data reduction were quantified for two separate classification methods, linear discriminant analysis (LDA) and support vector machine (SVM). As the model dataset, reflection data were acquired using a hyperspectral camera in 230 spectral channels from 401 to 879 nm (spectral resolution of 2.1 nm) from field pea (Pisum sativum) samples with and without internal pea weevil (Bruchus pisorum) infestation. We deployed five levels of spatial data reduction (binning) and eight levels of spectral data reduction (40 datasets). Forward stepwise LDA was used to select and include only spectral channels contributing the most to the separation of pixels from non-infested and infested field peas. Classification accuracies obtained with LDA and SVM were based on the classification of independent validation datasets. Overall, SVMs had significantly higher classification accuracies than LDAs (P < 0.01). There was a negative association between pixel resolution and classification accuracy, while spectral binning equivalent to up to 98% data reduction had negligible effect on classification accuracies. This study supports the potential use of reflection-based technologies in the quality control of food products with internal defects, and it highlights that spatial and spectral data reductions can (1) improve classification accuracies, (2) vastly decrease computer constraints, and (3) reduce analytical concerns associated with classifications of large and high-dimensional datasets.

KW - Field pea

KW - Hyperspectral imaging

KW - Machine vision

KW - Pea weevil

KW - Reflection data

KW - Spatial resolution

KW - Spectral resolution

UR - http://www.scopus.com/inward/record.url?scp=84924080926&partnerID=8YFLogxK

U2 - 10.1366/14-07672

DO - 10.1366/14-07672

M3 - Article

VL - 69

SP - 473

EP - 480

JO - Applied Spectroscopy

T2 - Applied Spectroscopy

JF - Applied Spectroscopy

SN - 0003-7028

IS - 4

ER -