TY - GEN
T1 - Authentication of premium tea based on geographical origin using NIR spectroscopy and multivariate analysis
AU - Hammel Sobreira, C.
AU - da Silva Ferreira, M. V.
AU - Kamruzzaman, M.
N1 - Publisher Copyright:
© 2023 ASABE Annual International Meeting. All Rights Reserved.
PY - 2023
Y1 - 2023
N2 - The authentication of the origin of the maker of the tea is extremely important to guarantee the quality of this plantation crop where near-infrared (NIR) spectroscopy stands out as a better alternative than laborious and expensive methods, such as gas chromatography-mass spectroscopy (CG-MS) and high-pressure liquid chromatography (HPLC) to help to discriminate the origin between samples. Thus, this work aimed to investigate a low-cost portable and benchtop NIR to classify tea leaves from two different origins (Brazil and India). Approximately 260 tea samples from Brazil (60%) and India (40%) were used for the investigation. PCA for portable NIR (PNIR) and benchtop (BNIR) showed a complete separation between the two groups. LDA and PLS-DA were used to classify the tea leaves into their respective origins. For PLS-DA, full spectra were used and the samples were divided into 70% for calibration and the remaining for validation. LDA and PLS-DA showed an accuracy of over 90% for both devices. Therefore, both low-cost PNIR, despite having different machinery (Fourier transformation feature) and BNIR, was able to classify the two classes of tea with reasonable accuracy. However, the former is easier to manage and cheaper than the latter and there is no need for a lab to do such a test. Thus, the correct distinction between these tea origins may prevent adding less expensive (mostly Brazilian) teas to Indian tea in the market.
AB - The authentication of the origin of the maker of the tea is extremely important to guarantee the quality of this plantation crop where near-infrared (NIR) spectroscopy stands out as a better alternative than laborious and expensive methods, such as gas chromatography-mass spectroscopy (CG-MS) and high-pressure liquid chromatography (HPLC) to help to discriminate the origin between samples. Thus, this work aimed to investigate a low-cost portable and benchtop NIR to classify tea leaves from two different origins (Brazil and India). Approximately 260 tea samples from Brazil (60%) and India (40%) were used for the investigation. PCA for portable NIR (PNIR) and benchtop (BNIR) showed a complete separation between the two groups. LDA and PLS-DA were used to classify the tea leaves into their respective origins. For PLS-DA, full spectra were used and the samples were divided into 70% for calibration and the remaining for validation. LDA and PLS-DA showed an accuracy of over 90% for both devices. Therefore, both low-cost PNIR, despite having different machinery (Fourier transformation feature) and BNIR, was able to classify the two classes of tea with reasonable accuracy. However, the former is easier to manage and cheaper than the latter and there is no need for a lab to do such a test. Thus, the correct distinction between these tea origins may prevent adding less expensive (mostly Brazilian) teas to Indian tea in the market.
KW - PLS regression, discrimination analysis
KW - black tea
KW - multivariate analysis
UR - http://www.scopus.com/inward/record.url?scp=85183588920&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85183588920&partnerID=8YFLogxK
U2 - 10.13031/aim.202300636
DO - 10.13031/aim.202300636
M3 - Conference contribution
AN - SCOPUS:85183588920
T3 - 2023 ASABE Annual International Meeting
BT - 2023 ASABE Annual International Meeting
PB - American Society of Agricultural and Biological Engineers
T2 - 2023 American Society of Agricultural and Biological Engineers Annual International Meeting, ASABE 2023
Y2 - 9 July 2023 through 12 July 2023
ER -