The phenolic compound concentration of olives and olive oil is typically quantified using HPLC; however, this process is expensive and time consuming. The purpose of this work was to evaluate the potential of Fourier transform infrared (FTIR) spectroscopy combined with chemometrics, as a rapid tool for the quantitative prediction of phenol content and antioxidant activity in olive fruits and oils from “Cobrançosa” cultivar. Normalized spectral data using standard normal variate (SNV) and first and second Savitzky–Golay derivatives were used to build calibration models based on principal component regression (PCR) and on partial least squares regression (PLS-R), the performance of both models have been also compared. It was shown the possibility of establishing optimized regression models using the combined frequency regions of 3050–2750 and 1800–790 cm−1 instead of the full mid-infrared spectrum was shown. It was concluded that, in general, the first derivative of data and PLS-R models offered enhanced results. Low root-mean-square error (RMSE) and high correlation coefficients (R2) for the calibration and for the validation sets were obtained.
- Chemical composition
- Olive fruit antioxidant activity
- Olive oil antioxidant activity
- Ripening stage