Seedlessness detection in White Malaga table grapes using near-infrared spectroscopy

Kittiwachana S., . and Nakano K., . and Ohashi S., . and Maniwara P., . and Kanchanomai C., . and Theanjumpol P., . and Krongchai C., . and Naphrom D., . Seedlessness detection in White Malaga table grapes using near-infrared spectroscopy. pp. 806-813. ISSN 22317546

White Malaga table grapes are seeded and widely grown in Thailand. They are converted by induction into seedless grapes to increase their value. It is difficult to identify seedlessness in table grapes without destroying the grape berry. The present work thus described a quick and non-destructive method for detecting and predicting seedlessness in White Malaga table grapes by using near-infrared (NIR) spectroscopy together with chemometric analysis. The NIR spectra of 280 grape samples were recorded after harvest. Firmness total soluble solids (TSS) pH titratable acidity (TA) tartaric acid number of seeds and relevant physical properties were analysed. The width and weight of plant growth regulator (PGR) treatments were significantly lower than those in the untreated grapes while the length firmness TA and tartaric acid were not significantly different. Partial least square (PLS) regression was used to investigate the prediction. Classification models namely principal component analysis (PCA) and quadratic discriminant analysis (QDA) were used to identify seedlessness. It was found that QDA as a representative of linear classification resulted in the best classification of seeded and seedless performance where the percentages of predictive ability (PA) the percentages of model stability (MS) and the percentages of correctly classified (CC) were 97.27 98.57 and 96.23 respectively for the training set with no pre-processing. Therefore the NIR spectroscopy technique can be a non-destructive technique for seedlessness detection in White Malaga table grapes.

Full text available from: Official URL: http://www.ifrj.upm.edu.my/29%20(04)%202022/08%20-...

Abstract

White Malaga table grapes are seeded and widely grown in Thailand. They are converted by induction into seedless grapes to increase their value. It is difficult to identify seedlessness in table grapes without destroying the grape berry. The present work thus described a quick and non-destructive method for detecting and predicting seedlessness in White Malaga table grapes by using near-infrared (NIR) spectroscopy together with chemometric analysis. The NIR spectra of 280 grape samples were recorded after harvest. Firmness total soluble solids (TSS) pH titratable acidity (TA) tartaric acid number of seeds and relevant physical properties were analysed. The width and weight of plant growth regulator (PGR) treatments were significantly lower than those in the untreated grapes while the length firmness TA and tartaric acid were not significantly different. Partial least square (PLS) regression was used to investigate the prediction. Classification models namely principal component analysis (PCA) and quadratic discriminant analysis (QDA) were used to identify seedlessness. It was found that QDA as a representative of linear classification resulted in the best classification of seeded and seedless performance where the percentages of predictive ability (PA) the percentages of model stability (MS) and the percentages of correctly classified (CC) were 97.27 98.57 and 96.23 respectively for the training set with no pre-processing. Therefore the NIR spectroscopy technique can be a non-destructive technique for seedlessness detection in White Malaga table grapes.

[error in script]

Item Type:	Article
AGROVOC Term:	Grapes
AGROVOC Term:	Near infrared spectroscopy
AGROVOC Term:	Gibberellic acid
AGROVOC Term:	Seedlessness
AGROVOC Term:	Seeds
AGROVOC Term:	Measurement
AGROVOC Term:	Physical properties
AGROVOC Term:	Chemical properties
AGROVOC Term:	Firmness
AGROVOC Term:	Tartaric acid
Depositing User:	Mr. AFANDI ABDUL MALEK
Last Modified:	24 Apr 2025 00:55
URI:	http://webagris.upm.edu.my/id/eprint/10822

Actions (login required)

View Item