Yield Predictions of Four Hybrids of Maize (Zea mays) Using Multispectral Images Obtained from UAV in the Coast of Peru

Early assessment of crop development is a key aspect precision agriculture. Shortening the time response before deficit irrigation, nutrients and damage by diseases one usual concerns in prediction yields can increase profitability for farmer’s economy. In this study, we aimed to predict yield four maize commercial hybrids (Dekalb7508, Advanta9313, MH_INIA619 Exp_05PMLM) using vegetation indices (VIs). A total 10 VIs (NDVI, GNDVI, GCI, RVI, NDRE, CIRE, CVI, MCARI, SAVI, CCCI) were considered evaluating plant cover at 31, 39, 42, 46 51 days after sowing (DAS). multivariate analysis was applied principal component (PCA), linear regression, r-Pearson correlation. Highly significant correlations found between with (GNDVI, CIRE DAS MCARI CCCI). The PCA showed clear discrimination dates evaluated 39 DAS. inclusion NDRE model contributed estimating performance, showing greater use unmanned aerial vehicles (UAVs) monitor crops allows us optimize resources helps making timely decisions agriculture Peru.