A Machine Learning Approach for the Non-Destructive Estimation of Leaf Area in Medicinal Orchid Dendrobium nobile L.

In this study, leaf area prediction models of Dendrobium nobile, were developed through machine learning (ML) techniques including multiple linear regression (MLR), support vector (SVR), gradient boosting (GBR), and artificial neural networks (ANNs). The best model was tested using the coefficient determination (R2), mean absolute errors (MAEs), root square (RMSEs) statistically confirmed average rank (AR). Leaf images captured a smartphone ImageJ used to calculate length (L), width (W), (LA). Three orders L, W, their combinations taken for building. Multicollinearity status checked Variance Inflation Factor (VIF) Tolerance (T). A total 80% dataset remaining 20% training validation, respectively. KFold (K = 10) cross-validation overfit. GBR (R2, MAE RMSE values ranged at 0.96, (0.82–0.91) (1.10–1.11) cm2) in testing phase among ML models. AR confirms outperformance GBR, securing first frequency top ten Thus, is imparting its future utilization estimate D. nobile.