Ripeness Evaluation of Achacha Fruit Using Hyperspectral Image Data

In this study, spectral data within the wavelength range of 400–780 nm were used to evaluate ripeness stages achacha fruits. The status fruits was divided into seven stages. Both average and pixel-based approaches assess ripeness. accuracy n-level-error each stage predicted by using classification models (Support Vector Machine (SVM), Partial Least Square Discriminant Analysis (PLS-DA), Artificial Neural Network (ANN) K-Nearest Neighbor (KNN)) regression (Partial Regression (PLSR) Support (SVR)). Furthermore, how curvature fruit surface affected prediction investigated. With use an averaged spectrum samples, model in study ranged from 52.25% 79.75%, one-level error (94.75–100%) much higher. SVM had highest (79.75%), PLSR (100%). results majority rule, (58.25–79.50%) one-level-error (95.25–99.75%) all comparable with spectrum. showed that could have a noticeable effect on evaluation values low or high stage. Thus, central region would be relatively reliable choice for evaluation. For fruit, value face exposed sunlight one level higher than shadow. when close mid-value two adjacent values, chance having errors. sorting practical postharvest processing