Evaluation of a robust regression method (RoBoost-PLSR) to predict biochemical variables for agronomic applications: Case study of grape berry maturity monitoring

Visible and near infrared spectroscopy (VIS-NIR) is increasingly being transferred from laboratory to industry for in-line portable applications in various domains. By intensively using VIS-NIR spectroscopy, some abnormal observations may certainly arise. It then important properly handle outliers elaborate effective prediction models. The objective of this study investigate the potential a robust method called Roboost-PLSR improve model performances viticulture application. This work focuses on case predict sugar content grape berries three different varieties Vitis Vinifera maturity monitoring context. Hyperspectral images were acquired Syrah, Fer-Servadou Mauzac varieties. Reference measurements levels made by densimetric baths. Performances RoBoost-PLSR models compared reference Partial Least Square Regression (PLSR). criteria PLSR obtained all with these following values: Syrah (Rp2 ?= ?0.971; RMSEp ?5.36 ?g/L), Fer-servadou ?0.788; ?11.69 ?g/L) ?0.690; ?15.61 ?g/L). Prediction qualities are improved RoBoost-PLSR: ?0.990; ?3.14 ?0.848; ?10.20 ?0.927; ?7.58 Results confirm that allows better consideration within calibration set.