Machine-Learning Approaches in N Estimations of Fig Cultivations Based on Satellite-Born Vegetation Indices
Karla Janeth Martínez-Macias,
Aldo Rafael Martínez-Sifuentes,
Selenne Yuridia Márquez-Guerrero,
Arturo Reyes-González,
Pablo Preciado-Rangel,
Pablo Yescas-Coronado,
Ramón Trucíos-Caciano
Affiliations
Karla Janeth Martínez-Macias
Research and Postgraduate Studies Division, Technological Institute of Torreon, Torreon 27170, Mexico
Aldo Rafael Martínez-Sifuentes
National Institute of Forestry, Agricultural and Livestock Research (INIFAP), National Center for Disciplinary Research on Water, Soil, Plant and Atmosphere Relationships (CENID-RASPA), Gomez Palacio 35150, Mexico
Selenne Yuridia Márquez-Guerrero
Research and Postgraduate Studies Division, Technological Institute of Torreon, Torreon 27170, Mexico
Arturo Reyes-González
National Institute of Forestry, Agricultural and Livestock Research (INIFAP), Experimental Field La Laguna, Matamoros 27440, Mexico
Pablo Preciado-Rangel
Research and Postgraduate Studies Division, Technological Institute of Torreon, Torreon 27170, Mexico
Pablo Yescas-Coronado
Research and Postgraduate Studies Division, Technological Institute of Torreon, Torreon 27170, Mexico
Ramón Trucíos-Caciano
National Institute of Forestry, Agricultural and Livestock Research (INIFAP), National Center for Disciplinary Research on Water, Soil, Plant and Atmosphere Relationships (CENID-RASPA), Gomez Palacio 35150, Mexico
Nitrogen is one of the most important macronutrients for crops, and, in conjunction with artificial intelligence algorithms, it is possible to estimate it with the aid of vegetation indices through remote sensing. Various indices were calculated and those with a correlation of ≥0.7 were selected for subsequent use in random forest, gradient boosting, and artificial neural networks to determine their relationship with nitrogen levels measured in the laboratory. Random forest showed no relationship, yielding an R2 of zero; and gradient boosting and the classical method were similar with 0.7; whereas artificial neural networks yielded the best results with an R2 of 0.93. Thus, estimating nitrogen levels using this algorithm is reliable, by feeding it with data from the Modified Chlorophyll Absorption Ratio Index, Transformed Chlorophyll Absorption Reflectance Index, Modified Chlorophyll Absorption Ratio Index/Optimized Soil Adjusted Vegetation Index, and Transformed Chlorophyll Absorption Ratio Index/Optimized Soil Adjusted Vegetation Index
