Estimation of strawberry firmness using hyperspectral imaging: a comparison of regression models

Firmness is one of the most important quality measures strawberries, and related to other aspects fruit, such as flavour, ripeness internal characteristics. The popular method for measuring firmness puncturing with a penetrometer, which destructive time-consuming. In present study, we make an attempt predict strawberries in fast, non-destructive non-contact way using hyperspectral imaging (HSI) data analysis various regression techniques. primary goal this research investigate compare prediction capability seven prominent We have performed HSI acquisition 150 optimised models spectral information strawberry firmness. These are linear, ridge, lasso, k-neighbours, random forest, support vector partial least square regression. res ults show that has potential rapid, manner. Out these models, k-neighbours model outperformed all methods standard error 0.14, better than state-of-the-art results.