A prototype on-line AOTF hyperspectral image acquisition system for tenderness assessment of beef carcasses

Govindarajan Konda Naganathan, Kim Cluff, Ashok K Samal, Chris R. Calkins, David D. Jones, Carol L. Lorenzen, Jeyamkondan Subbiah

Research output: Contribution to journalArticle

14 Citations (Scopus)

Abstract

A prototype on-line acousto-optic tunable filter (AOTF)-based hyperspectral image acquisition system (λ = 450-900 nm) was developed for tenderness assessment of beef carcasses. Hyperspectral images of ribeye muscle on stationary hanging beef carcasses (n = 338) at 2-day postmortem were acquired in commercial beef slaughter or packing plants. After image acquisition, a strip steak was cut from each carcass, vacuum packaged, aged for 14 days, cooked, and slice shear force tenderness scores were collected by an independent lab. Beef hyperspectral images were mosaicked together and principal component (PC) analysis was conducted to reduce the spectral dimension. Six different textural feature sets were extracted from the PC images and used in Fisher's linear discriminant model to classify beef samples into two tenderness categories: tender and tough. The pooled feature model performed better than the other models with a tender certification accuracy of 92.9% and 87.8% in cross-validation and third-party true validation, respectively. Two additional metrics namely overall accuracy and a custom defined metric called accuracy index, were used to compare the tenderness prediction models.

Original languageEnglish (US)
Pages (from-to)1-9
Number of pages9
JournalJournal of Food Engineering
Volume154
DOIs
StatePublished - Jun 2015

Fingerprint

hyperspectral imagery
beef carcasses
optics
prototypes
beef
packing houses
steaks
certification
shears
Certification
Vacuum
slaughter
Principal Component Analysis
principal component analysis
Linear Models
muscles
prediction
Red Meat
Muscles
sampling

Keywords

  • Acousto-optic tunable filter
  • Beef grading
  • Discriminant model
  • Feature selection
  • Principal component analysis
  • Textural features

ASJC Scopus subject areas

  • Food Science

Cite this

A prototype on-line AOTF hyperspectral image acquisition system for tenderness assessment of beef carcasses. / Konda Naganathan, Govindarajan; Cluff, Kim; Samal, Ashok K; Calkins, Chris R.; Jones, David D.; Lorenzen, Carol L.; Subbiah, Jeyamkondan.

In: Journal of Food Engineering, Vol. 154, 06.2015, p. 1-9.

Research output: Contribution to journalArticle

Konda Naganathan, Govindarajan ; Cluff, Kim ; Samal, Ashok K ; Calkins, Chris R. ; Jones, David D. ; Lorenzen, Carol L. ; Subbiah, Jeyamkondan. / A prototype on-line AOTF hyperspectral image acquisition system for tenderness assessment of beef carcasses. In: Journal of Food Engineering. 2015 ; Vol. 154. pp. 1-9.
@article{3ef11dee11c144788ea4dbc33a660add,
title = "A prototype on-line AOTF hyperspectral image acquisition system for tenderness assessment of beef carcasses",
abstract = "A prototype on-line acousto-optic tunable filter (AOTF)-based hyperspectral image acquisition system (λ = 450-900 nm) was developed for tenderness assessment of beef carcasses. Hyperspectral images of ribeye muscle on stationary hanging beef carcasses (n = 338) at 2-day postmortem were acquired in commercial beef slaughter or packing plants. After image acquisition, a strip steak was cut from each carcass, vacuum packaged, aged for 14 days, cooked, and slice shear force tenderness scores were collected by an independent lab. Beef hyperspectral images were mosaicked together and principal component (PC) analysis was conducted to reduce the spectral dimension. Six different textural feature sets were extracted from the PC images and used in Fisher's linear discriminant model to classify beef samples into two tenderness categories: tender and tough. The pooled feature model performed better than the other models with a tender certification accuracy of 92.9{\%} and 87.8{\%} in cross-validation and third-party true validation, respectively. Two additional metrics namely overall accuracy and a custom defined metric called accuracy index, were used to compare the tenderness prediction models.",
keywords = "Acousto-optic tunable filter, Beef grading, Discriminant model, Feature selection, Principal component analysis, Textural features",
author = "{Konda Naganathan}, Govindarajan and Kim Cluff and Samal, {Ashok K} and Calkins, {Chris R.} and Jones, {David D.} and Lorenzen, {Carol L.} and Jeyamkondan Subbiah",
year = "2015",
month = "6",
doi = "10.1016/j.jfoodeng.2014.12.015",
language = "English (US)",
volume = "154",
pages = "1--9",
journal = "Journal of Food Engineering",
issn = "0260-8774",
publisher = "Elsevier Limited",

}

TY - JOUR

T1 - A prototype on-line AOTF hyperspectral image acquisition system for tenderness assessment of beef carcasses

AU - Konda Naganathan, Govindarajan

AU - Cluff, Kim

AU - Samal, Ashok K

AU - Calkins, Chris R.

AU - Jones, David D.

AU - Lorenzen, Carol L.

AU - Subbiah, Jeyamkondan

PY - 2015/6

Y1 - 2015/6

N2 - A prototype on-line acousto-optic tunable filter (AOTF)-based hyperspectral image acquisition system (λ = 450-900 nm) was developed for tenderness assessment of beef carcasses. Hyperspectral images of ribeye muscle on stationary hanging beef carcasses (n = 338) at 2-day postmortem were acquired in commercial beef slaughter or packing plants. After image acquisition, a strip steak was cut from each carcass, vacuum packaged, aged for 14 days, cooked, and slice shear force tenderness scores were collected by an independent lab. Beef hyperspectral images were mosaicked together and principal component (PC) analysis was conducted to reduce the spectral dimension. Six different textural feature sets were extracted from the PC images and used in Fisher's linear discriminant model to classify beef samples into two tenderness categories: tender and tough. The pooled feature model performed better than the other models with a tender certification accuracy of 92.9% and 87.8% in cross-validation and third-party true validation, respectively. Two additional metrics namely overall accuracy and a custom defined metric called accuracy index, were used to compare the tenderness prediction models.

AB - A prototype on-line acousto-optic tunable filter (AOTF)-based hyperspectral image acquisition system (λ = 450-900 nm) was developed for tenderness assessment of beef carcasses. Hyperspectral images of ribeye muscle on stationary hanging beef carcasses (n = 338) at 2-day postmortem were acquired in commercial beef slaughter or packing plants. After image acquisition, a strip steak was cut from each carcass, vacuum packaged, aged for 14 days, cooked, and slice shear force tenderness scores were collected by an independent lab. Beef hyperspectral images were mosaicked together and principal component (PC) analysis was conducted to reduce the spectral dimension. Six different textural feature sets were extracted from the PC images and used in Fisher's linear discriminant model to classify beef samples into two tenderness categories: tender and tough. The pooled feature model performed better than the other models with a tender certification accuracy of 92.9% and 87.8% in cross-validation and third-party true validation, respectively. Two additional metrics namely overall accuracy and a custom defined metric called accuracy index, were used to compare the tenderness prediction models.

KW - Acousto-optic tunable filter

KW - Beef grading

KW - Discriminant model

KW - Feature selection

KW - Principal component analysis

KW - Textural features

UR - http://www.scopus.com/inward/record.url?scp=84920992700&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84920992700&partnerID=8YFLogxK

U2 - 10.1016/j.jfoodeng.2014.12.015

DO - 10.1016/j.jfoodeng.2014.12.015

M3 - Article

AN - SCOPUS:84920992700

VL - 154

SP - 1

EP - 9

JO - Journal of Food Engineering

JF - Journal of Food Engineering

SN - 0260-8774

ER -