Abstract

Near-infrared (NIR) spectroscopy has become a very popular technique for the non-invasive assessment of intact fruit. This work presents an application of a low-cost commercially available NIR spectrometer for the estimation of ripeness of Chilean wine grapes. Two configurations for the spectra acquisition were used (diffuse transmittance and interactance), using a custom-designed contact probe. Samples of Chardonnay, Carménère and Cabernet Sauvignon, collected over the 2002 harvest and pre-harvest seasons, were analysed for total soluble solids content (°Brix). Partial least squares calibration models, obtained from several preprocessing techniques (smoothing, multiplicative signal correction, standard normal variate, etc), were compared. Also, two spectral regions were used, one without the red part of the visible spectrum (just the short-wave (SW-NIR) region) and the other including it. Performance of different models was assessed in terms of root mean square of cross-validation, root mean square of prediction (RMSEP) and R2 for a validation set of samples. RMSEPs of 1.06 with R2 = 0.942 indicate that it is possible to estimate wine grape ripeness (°Brix value), by using a CCD portable spectrometer. The red grape models performed better than the white grape models.