Retrieval of Fractional Vegetation Cover from Remote Sensing Image of Unmanned Aerial Vehicle Based on Mixed Pixel Decomposition Method

FVC (fractional vegetation cover) is highly correlated with wheat plant density in the reviving period, which an important indicator for conducting variable-rate nitrogenous topdressing. In this study, objective of improving inversion accuracy density, innovative approach retrieval values from remote sensing images a UAV (unmanned aerial vehicle) was proposed based on mixed pixel decomposition method. Firstly, experimental field were acquired by using DJI Mini and endmembers image identified. Subsequently, linear unmixing model used to subdivide pixels into components soil, abundance map acquired. Based vegetation, calculated. Consequently, regression between ground truth data established. The coefficient determination (R2), RMSE (root mean square error), RRMSE (Relative-RMSE) calculated as 0.97, 1.86 plants/m2, 0.677%, indicates strong correlation method density. Therefore, we can conclude that significantly improved values, provides support basic fertilization period manner precision agriculture.