Integrating UAV Derived Vegetation and Texture Indices for Estimation of Leaf Nitrogen Concentration in Drip-Irrigated Cotton under Reduced Nitrogen Treatment and Different Plant Densities

Abstract

Abstract Background Accurate assessment of nitrogen (N) status is important for N management and yield improvement. The N status in plant was affected by plant densities and N application rates, while the methods of assessing the N status in drip-irrigated cotton under reduced nitrogen treatment and different plant densities are lacking. Methods This study was conducted with four different N treatments (195.5, 299, 402.5, and 506 kg N ha− 1) and three sowing densities (6.9×104, 13.8×104, and 24×104 plants ha− 1) by using a low-cost Unmanned Aerial Vehicle (UAV) system to acquire RGB imagery at 10 m flight altitude at cotton main growth stages. We evaluated the performance of different ground resolutions (1.3-, 2.6-, 5.2-,10.4-, 20.8-, 41.6-, 83.2-, and 166.4-cm-ground-resolution) image textures, vegetation indices (VIs), and their combination for leaf N concentrations (LNC) estimation with four regression methods (stepwise multiple linear regression, SMLR; support vector regression, SVR; extreme learning machine, ELM; random forest, RF). Results The results showed that the combination of VIs and texture maintained higher estimation accuracy than using VIs or textures alone. Specifically, the RF regression models had the higher accuracy and stability than SMLR and other two machine learning algorithms. The best accuracy (R2 = 0.87, RMSE = 3.14g kg− 1, rRMSE = 7.00%) was obtained when RF was applied in combination with VIs and texture. Conclusion The combination of VIs and textures from UAV images using RF could improve the estimation accuracy of drip-irrigated cotton LNC and may have the potential contribution in rapid and non-destructive nutrition monitoring and diagnosis of other crops or other growth parameters.