Use of Artificial Neural Networks and NIR Spectroscopy for Non-Destructive Grape Texture Prediction

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 (R-2 0.82 and 0.72). Qualitative models were obtained for hardness and chewiness (R-2 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.

Automated summary

In this study, non-destructive NIR spectroscopy and machine learning techniques were used to predict texture parameters and total soluble solids content in intact grape berries. The multivariate models built using artificial neural networks and partial least squares regressions showed improved prediction accuracy after eliminating uninformative spectral ranges. A high prediction accuracy was achieved for total soluble solids content and springiness, while qualitative models were obtained for hardness and chewiness. Unfortunately, a satisfactory calibration model could not be established for cohesiveness. The size of the grape berries is closely related to their textural parameters, requiring a time-consuming sorting step before texture measurement.