Neuro-Fuzzy System to Predict Timely Harvest in Stevia Crops DOI Creative Commons

Shanti-Maryse Gutiérrez-Magaña,

Noel García-Díaz, Leonel Soriano-Equigua

и другие.

Agriculture, Год журнала: 2025, Номер 15(8), С. 840 - 840

Опубликована: Апрель 13, 2025

Agriculture is essential for food production and raw materials. A key aspect of this sector harvest, the stage at which commercial part plant separated. Timely harvesting minimizes post-harvest losses, preserves product quality, optimizes processes. Globally, a substantial amount wasted, impacting security natural resources. To address problem, an Adaptive Neuro-Fuzzy Inference System was developed to predict timely in crops. Stevia, native from Brazil Paraguay, with annual 100,000 200,000 tons market 400 million dollars, focus study. The system considers soil pH, Brix Degrees, leaf colorimetry as inputs. output binary: 1 indicates harvest 0 no harvest. assess its performance, Leave-One-Out Cross-Validation used, obtaining r2 0.99965 Absolute Residual Error 0.00064305, demonstrating accuracy robustness. In addition, interactive application that allows farmers evaluate crop status optimize decision-making developed.

Язык: Английский

Neuro-Fuzzy System to Predict Timely Harvest in Stevia Crops DOI Creative Commons

Shanti-Maryse Gutiérrez-Magaña,

Noel García-Díaz, Leonel Soriano-Equigua

и другие.

Agriculture, Год журнала: 2025, Номер 15(8), С. 840 - 840

Опубликована: Апрель 13, 2025

Agriculture is essential for food production and raw materials. A key aspect of this sector harvest, the stage at which commercial part plant separated. Timely harvesting minimizes post-harvest losses, preserves product quality, optimizes processes. Globally, a substantial amount wasted, impacting security natural resources. To address problem, an Adaptive Neuro-Fuzzy Inference System was developed to predict timely in crops. Stevia, native from Brazil Paraguay, with annual 100,000 200,000 tons market 400 million dollars, focus study. The system considers soil pH, Brix Degrees, leaf colorimetry as inputs. output binary: 1 indicates harvest 0 no harvest. assess its performance, Leave-One-Out Cross-Validation used, obtaining r2 0.99965 Absolute Residual Error 0.00064305, demonstrating accuracy robustness. In addition, interactive application that allows farmers evaluate crop status optimize decision-making developed.

Язык: Английский

Процитировано

0