Rapid Prediction of Protein Content in Chickpeas Using Hyperspectral Imaging

Original Title: Application of near-infrared hyperspectral imaging coupled with chemometrics for rapid and non-destructive prediction of protein content in single chickpea seed

Saha et al. Application of near-infrared hyperspectral imaging coupled with chemometrics for rapid and non-destructive prediction of protein content in single chickpea seed Volume 115, January 2023,104939 https://doi.org/10.1016/j.jfca.2022.104938

Assessing the protein content of a single chickpea seed quickly and non-destructively is crucial for developing high-protein varieties. Due to this, in this study the use of near-infrared hyperspectral imaging (NIR-HSI) were studied.

Eight chickpea varieties were analyzed within the 900–2500 nm range, measuring the seed in two positions: micropyle up and micropyle down. Spectral data were used to build predictive models using partial least squares regression and support vector machine regression. To enhance accuracy, spectral preprocessing techniques were applied.

The best model reached, in prediction, a correlation coefficient of 0.947 and a root mean square error of 0.861.

These findings demonstrate that NIR hyperspectral imaging, combined with advanced data analysis, enables efficient protein content prediction in chickpeas, supporting the breeding of more nutritious varieties.

Would you like to know more? Check it out Journal of Food Composition and Analysis, Elsevier