基于无人机遥感的企业苜蓿产量估测研究OA

Alfalfa(Medicago sativa)occupies an important position in global animal husbandry due to its high yield and rich nutritional value.With the rapid development of agricultural information technology,the innovative application of unmanned aerial vehicle(UAV)remote sensing technology provides a new multifunctional technical means for enterprises to monitor the growth dynamics of alfalfa.This study designed irrigation,fertilization,and water-fertilizer interaction experiments at the production base of the enterprise.UAV multispectral data were collected for three growth periods,extracting 14 vegetation indices(VI)and 40 texture features(TF).Multiple linear regression(MLR),random forest(RF),and support vector machine(SVM)models were used to construct alfalfa yield estimation models.The research findings indicated that:1)The average canopy reflectance of alfalfa was highest under a gradient of medium-high water and high fertilizer.2)The normalized difference red edge index,vegetation stress index,and mean values of green-and red-light bands exhibited the most significant correlation with yield.The synergistic effects of water and fertilizer effectively enhanced the correlation between VI,TF,and yield(P<0.05).3)The combination of VI and TF achieved the best estimation results in a RF model,with an R2 of 0.89,low error,and no overfitting.