In this article, a combination of non-destructive NIR spectroscopy and machine learning techniques was applied to predict the texture parameters and the total soluble solids content (TSS) in intact berries. The multivariate models obtained by building artificial neural networks (ANNs) and applying partial least squares (PLS) regressions showed a better prediction ability after the elimination of uninformative spectral ranges. A very good prediction was obtained for TSS and springiness (R² 0.82 and 0.72). Qualitative models were obtained for hardness and chewiness (R² 0.50 and 0.53). No satisfactory calibration model could be established between the NIR spectra and cohesiveness. Textural parameters of grape are strictly related to the berry size. Before any grape textural measurement, a time-consuming berry-sorting step is compulsory. This is the first time a complete textural analysis of intact grape berries has been performed by NIR spectroscopy without any a priori knowledge of the berry density class.

Texture characteristics are valuable parameters in the perceived quality and overall acceptability of fresh fruit. The characterization of grape texture attributes, such as firmness and crunchiness, is usually performed by sensory analysis or instrumental texture analysis. Both methodologies are destructive. Hence, it is not possible to test multiple times or perform any other analysis on the same sample. In this article, near-infrared (NIR) spectroscopy was applied to intact berries of table grape cv. Regal Seedless. NIR spectra were employed to predict both the physical parameter “hardness”, which is correlated with the crunchiness of berry flesh and the sweetness, which is correlated with the total soluble solids content (TSS, as °Brix). The chemometric analysis was carried out exclusively based on an open-source software environment, producing results readily usable for any operator, besides the specific level of experience with NIR spectroscopy.