DefinePK

Abstract

The present study aimed at to follow the tomato fruit development and quality by hand-held near-infrared spectroscopy. Tomato quality were followed from few days after fruit setting until harvest at commercial maturity during two seasons (spring and summer).
Results showed that in both seasons, fruit can be classified from fruit setting to harvest at maturity by using qualitative models (factorial discriminant analyses).
Quantitatives models based on PLS regressions allowed the prediction of soluble solids content (R=0.9, RMSE=0.1%Brix), titrable acidity (R=0.9, RMSE=0.6meq.100g-1) and color (a*, R=0.9, RMSE=5) of fruit. The accuracy of the predictions depend on the season and also on the maturity stage.
the results are promising in the context of developing a tool to assist in fruit phenotyping on site. Other experiment are now necessary to improve the accuracy and the robustness of the models with including additional varieties growing under variable climatic conditions in our greenhouses.