Evaluation of the ripening stages of apple (Golden Delicious) by means of computer vision system

Mexican apple production suffers high losses due to poor handling in processing. Implementation of a straightforward, low cost method to sort apples by their ripening stage is required. A set of Golden Delicious apples was used to monitor their physicochemical properties and external colour, a second set of apples was used to validate the method. To classify the stages, a ripening index (RPI) was proposed, in which three stages were identified; unripe, ripe and senescent. Weibull model was applied to the physicochemical parameters in order to describe their kinetic behaviour. The three RPI stages were compared with colour variability using the CIELab colour space, chroma (C∗) and hue angle (h∗), allowing the identification of the three ripening stages. Principal component analysis was used to evaluate the correlation between variables. A first correlation was performed between physicochemical and colour parameters and variables correlated correctly between each other except for L∗, but both described the samples variability with 91.05% reliability. Using only colour parameters, the samples were described accurately with 95.06% reliability. Multivariate discriminant analysis (MDA) was done in order to validate the method. A cross-validation was performed with an initial set of apples used as trial samples and a second set of apples for validation. MDA was capable of classifying apples in their correct ripening stage with 100% accuracy. A second analysis was carried out using four colour parameters (a∗, b∗, C and h∗), and results indicated that the ripening stages can be classified with 100% accuracy.