Improving estimation of LAI dynamic by fusion of morphological and vegetation indices based on UAV imagery

As an important indicator reflecting plant growth and canopy structure, accurate and rapid monitoring of the leaf area index (LAI) is very important for modern precision agriculture. The purpose of this study is to explore the potential of fusion of morphological information and spectral information in multiple growth periods of maize to improve the accuracy of LAI dynamic estimation. The multi-spectral sensor carried by the unmanned aerial vehicle (UAV) was used to collect remote sensing images of the maize canopy during the six growth stages. Three morphological parameters (canopy height, canopy coverage, and canopy volume) and two vegetation indices (normalized vegetation index (NDVI) and visible atmospheric vegetation index (VARI)) were extracted from image information and spectral information, respectively, and a LAI estimation model was constructed based on parameters fusion. The results showed that the morphological parameters and vegetation indices had the same time distribution law as LAI, and could be used to monitor crop LAI. At the same time, the study found that the fusion of canopy height, canopy coverage and canopy volume could further characterize the external morphological changes of crops and improved the accuracy of LAI dynamic estimation based on morphology, but there were still limitations in the seedling and milk stages. Furthermore, the fusion of canopy morphological parameters and vegetation indices could further improve the dynamic estimate accuracy of maize LAI, and showed better performance in all growth stages (Seedling stage: Rv(2) = 0.688, RMSEP = 0.0493; Jointing stage: Rv(2) = 0.860, RMSEP = 0.0847; Tasseling stage: Rv(2) = 0.780, RMSEP = 0.1829; Silking stage: Rv(2) = 0.794, RMSEP = 0.1981; Blister stage: Rv(2) = 0.793, RMSEP = 0.1584; Milk stage: Rv(2) = 0.708, RMSEP = 0.1396; All: Rv(2) = 0.943, RMSEP = 0.2618). The results show that the fusion of image information and spectral information can improve the estimation accuracy of crop LAI and provide a feasible method for crop growth information monitoring based on UAV platform.

Automated summary

This study explores the potential of fusing morphological information and spectral information to improve the accuracy of leaf area index (LAI) estimation for maize in multiple growth stages. The results show that the fusion of canopy morphological parameters and vegetation indices can improve the dynamic estimate accuracy of maize LAI and provide a feasible method for crop growth information monitoring based on UAV platform. The study highlights the importance of accurately and rapidly monitoring LAI for precision agriculture.