Skip to main content
ARS Home » Pacific West Area » Albany, California » Western Regional Research Center » Healthy Processed Foods Research » Research » Publications at this Location » Publication #374942

Research Project: Defining, Measuring, and Mitigating Attributes that Adversely Impact the Quality and Marketability of Foods

Location: Healthy Processed Foods Research

Title: Pine nut species recognition using NIR spectroscopy and image analysis

Author
item MOSCETTI, ROBERTO - University Of Tuscia
item BERHE, DANIEL - University Of Tuscia
item AGRIMI, MARIAGRAZIA - University Of Tuscia
item Haff, Ronald - Ron
item Liang, Peishih
item FERRI, SERENA - University Of Tuscia
item MONARCA, DANILO - University Of Tuscia
item MASSANTINI, RICCARDO - University Of Tuscia

Submitted to: Food and Bioprocess Technology
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 9/20/2020
Publication Date: 9/21/2020
Citation: Moscetti, R., Berhe, D.H., Agrimi, M., Haff, R.P., Liang, P., Ferri, S., Monarca, D., Massantini, R. 2020. Pine nut species recognition using NIR spectroscopy and image analysis. Food and Bioprocess Technology. 292. https://doi.org/10.1016/j.jfoodeng.2020.110357.
DOI: https://doi.org/10.1016/j.jfoodeng.2020.110357

Interpretive Summary: NIR spectroscopy and physical properties derived from image analysis were evaluated as potential data for statistical algorithms to distinguish seed kernels from two pine nut species (P. pinea and P. sibirica). Model performances were evaluated in terms of specificity, sensitivity and accuracy. Data pre-treatments were essential for achieving excellent performances (accuracy rate > 95%) in all tests. The interval algorithms indicated that the most important features for (1) the NIR method were the absorption bands at 1640-1658, 1720-1738 and 1880-1998 nm, while for (2) the image analysis were kernel eccentricity, kernel major axis length, kernel luminance (L*) and kernel perimeter. The results demonstrate potential for both techniques for discriminating pine nut species.

Technical Abstract: NIR spectroscopy and physical properties derived from image analysis were evaluated as potential features for the classification of seed kernels from two pine nut species (P. pinea and P. sibirica) using Partial Least Squares Discriminant Analysis (PLS-DA). Model performances were evaluated in terms of specificity, sensitivity and accuracy. Data pre-treatments were essential for achieving excellent performances (accuracy rate > 95%) in all tests. The interval PLS-DA highlighted that the most important features for (1) the NIR method were the absorption bands at 1640-1658, 1720-1738 and 1880-1998 nm, while for (2) the image analysis were kernel eccentricity, kernel major axis length, kernel luminance (L*) and kernel perimeter. The results demonstrate potential for both techniques for discriminating pine nut species.