A Novel Approach for Non-Invasive Grading and Sorting of Mango Fruit (Mangifera Indica L.) With Soil Quality Using ant Lion Optimizer (ALO) Artificial Neural Networks

Resumen

Mangoes are one of the most significant commercial crops in terms of market value and production volume, with their extensive production in over 90 countries worldwide. The demand for efficient grading and sorting processes has escalated due to the increasing popularity of mangoes in both domestic and international markets. The current study examines a new and effective non-invasive grading and sorting model for mangoes implemented using a hybrid soft computing approach. The system employs Artificial Neural Networks (ANN) optimized with Ant Lion Optimizer (ALO) as a classification tool, which evaluates the mangoes based on four essential grading parameters, including size (volume and morphology), shape, color, maturity (ripe/unripe), defect (defective/healthy), and variety (cultivar). The study has demonstrated that the proposed model achieves an overall classification rate of 96.82% and outperforms the other models. In addition, the study has conducted various experiments to validate the effectiveness of the proposed model and has compared it with existing grading and sorting models. The findings of this study highlight the significance of non-invasive methods in the grading and sorting of mangoes and the promising potential of hybrid soft computing approaches in improving the efficiency and accuracy of the process.