| Sumario: | Progression in the technological facets such as, remote sensing, machine learning, big data analytics, unmanned aerial vehicles (UAV’s) etc. have unraveled a new domain of sensor-based, non-invasive, fast, and near-real time assessment of crop conditions for efficient site-specific management. Present study attempts to map plant nitrogen (N) for rice crop in spatial scale in the experimental farm of ICAR-Indian Agricultural Research Institute (IARI), New Delhi, India using unmanned aerial vehicle (UAV) integrated with multispectral sensor. Five different treatments of nitrogen (No-N4 corresponding to 0, 50, 100, 150, and 200 kg N ha-1) were imposed to generate the variation in plant N content in the rice field. Four customary machine learning algorithms, namely, artificial neural network, support vector machines, random forest and multivariate adaptive regression splines regression ha red edge technology. The approach developed can be upscaled to the ve been implemented to retrieve N content. Evaluation of different machine learning algorithms led to development of optimized workflow for concurrent estimation of N content with reasonably good accuracy with the multispectral sensor using red edge technology. The approach developed can be upscaled to the farmer’s field to ensure feasible solutions for the prevailing challenges.
|
|---|