NIR and MIR spectroscopy to predict quality parameters in pink pepper (Schinus terebinthifolius Raddi) by chemometrics.

Conventional methods for food quality control are time-consuming and require large amounts of reagents. In this study, we present a novel application of near and mid-infrared spectroscopy combined with chemometrics to predict the physicochemical composition, phenolic constituents, (including total flavonoids, gallic acid, catechin, epicatechin, rutin, naringenin, procyanidin B1 and procyanidin B2) and various antioxidant parameters of pink pepper. Thirty-four models were optimized using regression by partial least squares (PLS and iPLS). The iPLS models showed excellent predictive capability with a minimum correlation coefficient of 0.7, low associated errors, and favorable RPD and RER values indicating practical applicability during model validation. Additionally, these models exhibited linearity with a slope close to 1 and an intercept near 0, presenting better fitting parameters than models generated with the full spectrum by PLS. These results suggest that spectroscopy can serve as an efficient alternative for routine quality analysis of pink pepper, enabling the rapid simultaneous prediction of 17 quality parameters while avoiding the laborious procedures associated with traditional chromatographic and spectrophotometric tests.