Advanced applications of near/mid-infrared (NIR/MIR) imaging spectroscopy for rapid prediction of potato and sweet potato moisture contents

Wen-Hao Su, Serafim Bakalis, Da-Wen Sun

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

Abstract

Near-infrared (NIR) and mid-infrared (MIR) hyperspectral techniques in tandem with chemometric analyses were employed for allowing the food industry to monitor moisture content (MC) in potato and sweet potato products during drying. Multivariate models were established by partial least-squares regression (PLSR), support vector machine regression (SVMR), locally weighted partial least square regression (LWPLSR), and back propagation artificial neural network (BPANN) using full spectral ranges of 10372-6105 cm-1 (Spectral Set ), 3996-600 cm-1 (Spectral Set ), and 1700-900 cm-1 (Spectral Set ). Non-linear methods in terms of LWPLSR and BPANN obtained better accuracies than linear models for tuber MC prediction. The overall model accuracy based on the NIR spectra (Spectral Set ) was higher than that of models using the MIR spectra. The highest accuracy (R2P = 0.987, RMSEP = 0.015) was achieved by the NIR-LWPLSR using Spectral Set . The results reveal that hyperspectral techniques have great potential to be used for measurement of tuber MC.

Original languageEnglish (US)
Title of host publicationASABE Annual International Meeting
DOIs
StatePublished - Jul 7 2019
Event2019 ASABE Annual International Meeting - Boston, United States
Duration: Jul 7 2019Jul 10 2019

Conference

Conference2019 ASABE Annual International Meeting
CountryUnited States
CityBoston
Period7/7/197/10/19

Fingerprint

Infrared imaging
sweet potatoes
least squares
spectroscopy
Moisture
image analysis
Spectroscopy
potatoes
Infrared radiation
water content
prediction
neural networks
tubers
Backpropagation
potato products
chemometrics
Neural networks
food industry
linear models
drying

Keywords

  • Drying
  • Moisture
  • Near-infrared
  • Potato
  • Spectral imaging

Cite this

Advanced applications of near/mid-infrared (NIR/MIR) imaging spectroscopy for rapid prediction of potato and sweet potato moisture contents. / Su, Wen-Hao; Bakalis, Serafim; Sun, Da-Wen.

ASABE Annual International Meeting. 2019.

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

Su, W-H, Bakalis, S & Sun, D-W 2019, Advanced applications of near/mid-infrared (NIR/MIR) imaging spectroscopy for rapid prediction of potato and sweet potato moisture contents. in ASABE Annual International Meeting. 2019 ASABE Annual International Meeting, Boston, United States, 7/7/19. https://doi.org/10.13031/aim.201900121
@inproceedings{de4f180eca114ae5bfb9e3e047f49767,
title = "Advanced applications of near/mid-infrared (NIR/MIR) imaging spectroscopy for rapid prediction of potato and sweet potato moisture contents",
abstract = "Near-infrared (NIR) and mid-infrared (MIR) hyperspectral techniques in tandem with chemometric analyses were employed for allowing the food industry to monitor moisture content (MC) in potato and sweet potato products during drying. Multivariate models were established by partial least-squares regression (PLSR), support vector machine regression (SVMR), locally weighted partial least square regression (LWPLSR), and back propagation artificial neural network (BPANN) using full spectral ranges of 10372-6105 cm-1 (Spectral Set ), 3996-600 cm-1 (Spectral Set ), and 1700-900 cm-1 (Spectral Set ). Non-linear methods in terms of LWPLSR and BPANN obtained better accuracies than linear models for tuber MC prediction. The overall model accuracy based on the NIR spectra (Spectral Set ) was higher than that of models using the MIR spectra. The highest accuracy (R2P = 0.987, RMSEP = 0.015) was achieved by the NIR-LWPLSR using Spectral Set . The results reveal that hyperspectral techniques have great potential to be used for measurement of tuber MC.",
keywords = "Drying, Moisture, Near-infrared, Potato, Spectral imaging",
author = "Wen-Hao Su and Serafim Bakalis and Da-Wen Sun",
year = "2019",
month = "7",
day = "7",
doi = "10.13031/aim.201900121",
language = "English (US)",
booktitle = "ASABE Annual International Meeting",

}

TY - GEN

T1 - Advanced applications of near/mid-infrared (NIR/MIR) imaging spectroscopy for rapid prediction of potato and sweet potato moisture contents

AU - Su, Wen-Hao

AU - Bakalis, Serafim

AU - Sun, Da-Wen

PY - 2019/7/7

Y1 - 2019/7/7

N2 - Near-infrared (NIR) and mid-infrared (MIR) hyperspectral techniques in tandem with chemometric analyses were employed for allowing the food industry to monitor moisture content (MC) in potato and sweet potato products during drying. Multivariate models were established by partial least-squares regression (PLSR), support vector machine regression (SVMR), locally weighted partial least square regression (LWPLSR), and back propagation artificial neural network (BPANN) using full spectral ranges of 10372-6105 cm-1 (Spectral Set ), 3996-600 cm-1 (Spectral Set ), and 1700-900 cm-1 (Spectral Set ). Non-linear methods in terms of LWPLSR and BPANN obtained better accuracies than linear models for tuber MC prediction. The overall model accuracy based on the NIR spectra (Spectral Set ) was higher than that of models using the MIR spectra. The highest accuracy (R2P = 0.987, RMSEP = 0.015) was achieved by the NIR-LWPLSR using Spectral Set . The results reveal that hyperspectral techniques have great potential to be used for measurement of tuber MC.

AB - Near-infrared (NIR) and mid-infrared (MIR) hyperspectral techniques in tandem with chemometric analyses were employed for allowing the food industry to monitor moisture content (MC) in potato and sweet potato products during drying. Multivariate models were established by partial least-squares regression (PLSR), support vector machine regression (SVMR), locally weighted partial least square regression (LWPLSR), and back propagation artificial neural network (BPANN) using full spectral ranges of 10372-6105 cm-1 (Spectral Set ), 3996-600 cm-1 (Spectral Set ), and 1700-900 cm-1 (Spectral Set ). Non-linear methods in terms of LWPLSR and BPANN obtained better accuracies than linear models for tuber MC prediction. The overall model accuracy based on the NIR spectra (Spectral Set ) was higher than that of models using the MIR spectra. The highest accuracy (R2P = 0.987, RMSEP = 0.015) was achieved by the NIR-LWPLSR using Spectral Set . The results reveal that hyperspectral techniques have great potential to be used for measurement of tuber MC.

KW - Drying

KW - Moisture

KW - Near-infrared

KW - Potato

KW - Spectral imaging

UR - http://www.scopus.com/inward/record.url?scp=85072911632&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85072911632&partnerID=8YFLogxK

U2 - 10.13031/aim.201900121

DO - 10.13031/aim.201900121

M3 - Conference contribution

BT - ASABE Annual International Meeting

ER -