Fusing spectral and textural information in near-infrared hyperspectral imaging to improve green tea classification modelling

Puneet Mishra, Alison Nordon, Mohd Shahrimie Mohd Asaari, Guoping Lian, Sally Redfern

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

Hyperspectral imaging (HSI) can acquire data in two modes: imaging and spectroscopy, revealing the spatially-resolved spectral properties of materials. Traditional HSI processing in the close-range domain primarily focuses on the spectral information with minimal utilisation of the spatial information present in the data. The present work describes a methodology for utilising the spatial information present in HSI data to improve classification modelling over that achievable with spectral information alone. The methodology has been evaluated using near infrared (NIR) HSI data of sixteen green tea products from seven different countries. The methodology involves selecting and sharpening an image plane to enhance the textural details. The textural information is then extracted from the statistical properties of the grey level co-occurrence matrix (GLCM) of the sharpened image plane using a moving window operation. Finally, the textural properties are combined with the spectral information using one of the three different levels of data fusion, i.e. raw data level, feature level and decision level. Raw data-level fusion involved concatenating the spectral and textural data before performing the classification task. The feature-level fusion involved performing principal component analysis (PCA) on spectral and textural information and combining the PC scores obtained prior to performing classification. Decision-level fusion involved a majority voting scheme to enhance the final classification maps. All the classification tasks were performed using multi-class support vector machine (SVM) models. The results showed that combining the textural and spectral information during modelling resulted in improved classification of the sixteen green tea products compared to models built using spectral or textural information alone.
LanguageEnglish
Pages40-47
Number of pages8
JournalJournal of Food Engineering
Volume249
Early online date19 Jan 2019
DOIs
Publication statusPublished - 1 May 2019

Fingerprint

green tea
Tea
image analysis
Politics
Principal Component Analysis
Spectrum Analysis
spectroscopy
principal component analysis
methodology

Keywords

  • chemical imaging
  • texture
  • support vector machine (SVM)
  • grey level co-occurrence matrix (GLCM)
  • data fusion
  • green tea

Cite this

@article{6ffb244dbd5b42228d975c1a4c266b8b,
title = "Fusing spectral and textural information in near-infrared hyperspectral imaging to improve green tea classification modelling",
abstract = "Hyperspectral imaging (HSI) can acquire data in two modes: imaging and spectroscopy, revealing the spatially-resolved spectral properties of materials. Traditional HSI processing in the close-range domain primarily focuses on the spectral information with minimal utilisation of the spatial information present in the data. The present work describes a methodology for utilising the spatial information present in HSI data to improve classification modelling over that achievable with spectral information alone. The methodology has been evaluated using near infrared (NIR) HSI data of sixteen green tea products from seven different countries. The methodology involves selecting and sharpening an image plane to enhance the textural details. The textural information is then extracted from the statistical properties of the grey level co-occurrence matrix (GLCM) of the sharpened image plane using a moving window operation. Finally, the textural properties are combined with the spectral information using one of the three different levels of data fusion, i.e. raw data level, feature level and decision level. Raw data-level fusion involved concatenating the spectral and textural data before performing the classification task. The feature-level fusion involved performing principal component analysis (PCA) on spectral and textural information and combining the PC scores obtained prior to performing classification. Decision-level fusion involved a majority voting scheme to enhance the final classification maps. All the classification tasks were performed using multi-class support vector machine (SVM) models. The results showed that combining the textural and spectral information during modelling resulted in improved classification of the sixteen green tea products compared to models built using spectral or textural information alone.",
keywords = "chemical imaging, texture, support vector machine (SVM), grey level co-occurrence matrix (GLCM), data fusion, green tea",
author = "Puneet Mishra and Alison Nordon and {Mohd Asaari}, {Mohd Shahrimie} and Guoping Lian and Sally Redfern",
year = "2019",
month = "5",
day = "1",
doi = "10.1016/j.jfoodeng.2019.01.009",
language = "English",
volume = "249",
pages = "40--47",
journal = "Journal of Food Engineering",
issn = "0260-8774",

}

Fusing spectral and textural information in near-infrared hyperspectral imaging to improve green tea classification modelling. / Mishra, Puneet; Nordon, Alison; Mohd Asaari, Mohd Shahrimie ; Lian, Guoping; Redfern, Sally .

In: Journal of Food Engineering, Vol. 249, 01.05.2019, p. 40-47.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Fusing spectral and textural information in near-infrared hyperspectral imaging to improve green tea classification modelling

AU - Mishra, Puneet

AU - Nordon, Alison

AU - Mohd Asaari, Mohd Shahrimie

AU - Lian, Guoping

AU - Redfern, Sally

PY - 2019/5/1

Y1 - 2019/5/1

N2 - Hyperspectral imaging (HSI) can acquire data in two modes: imaging and spectroscopy, revealing the spatially-resolved spectral properties of materials. Traditional HSI processing in the close-range domain primarily focuses on the spectral information with minimal utilisation of the spatial information present in the data. The present work describes a methodology for utilising the spatial information present in HSI data to improve classification modelling over that achievable with spectral information alone. The methodology has been evaluated using near infrared (NIR) HSI data of sixteen green tea products from seven different countries. The methodology involves selecting and sharpening an image plane to enhance the textural details. The textural information is then extracted from the statistical properties of the grey level co-occurrence matrix (GLCM) of the sharpened image plane using a moving window operation. Finally, the textural properties are combined with the spectral information using one of the three different levels of data fusion, i.e. raw data level, feature level and decision level. Raw data-level fusion involved concatenating the spectral and textural data before performing the classification task. The feature-level fusion involved performing principal component analysis (PCA) on spectral and textural information and combining the PC scores obtained prior to performing classification. Decision-level fusion involved a majority voting scheme to enhance the final classification maps. All the classification tasks were performed using multi-class support vector machine (SVM) models. The results showed that combining the textural and spectral information during modelling resulted in improved classification of the sixteen green tea products compared to models built using spectral or textural information alone.

AB - Hyperspectral imaging (HSI) can acquire data in two modes: imaging and spectroscopy, revealing the spatially-resolved spectral properties of materials. Traditional HSI processing in the close-range domain primarily focuses on the spectral information with minimal utilisation of the spatial information present in the data. The present work describes a methodology for utilising the spatial information present in HSI data to improve classification modelling over that achievable with spectral information alone. The methodology has been evaluated using near infrared (NIR) HSI data of sixteen green tea products from seven different countries. The methodology involves selecting and sharpening an image plane to enhance the textural details. The textural information is then extracted from the statistical properties of the grey level co-occurrence matrix (GLCM) of the sharpened image plane using a moving window operation. Finally, the textural properties are combined with the spectral information using one of the three different levels of data fusion, i.e. raw data level, feature level and decision level. Raw data-level fusion involved concatenating the spectral and textural data before performing the classification task. The feature-level fusion involved performing principal component analysis (PCA) on spectral and textural information and combining the PC scores obtained prior to performing classification. Decision-level fusion involved a majority voting scheme to enhance the final classification maps. All the classification tasks were performed using multi-class support vector machine (SVM) models. The results showed that combining the textural and spectral information during modelling resulted in improved classification of the sixteen green tea products compared to models built using spectral or textural information alone.

KW - chemical imaging

KW - texture

KW - support vector machine (SVM)

KW - grey level co-occurrence matrix (GLCM)

KW - data fusion

KW - green tea

UR - https://www.sciencedirect.com/journal/journal-of-food-engineering

U2 - 10.1016/j.jfoodeng.2019.01.009

DO - 10.1016/j.jfoodeng.2019.01.009

M3 - Article

VL - 249

SP - 40

EP - 47

JO - Journal of Food Engineering

T2 - Journal of Food Engineering

JF - Journal of Food Engineering

SN - 0260-8774

ER -