HEIGHT ESTIMATION OF MACAÚBA PLANTS USING DRONES AND DIGITAL ELEVATION MODELS

Macaúba crop is in the early stage of domestication, which requires a detailed understanding of its agronomic characteristics to maximize its use effectively. Estimating the height of the plants is one of the key points in this process. In particular, the height assessment helps to understand the adaptation of macaúba to the environment and genetic variability, as well as optimization in harvesting processes. To estimate the height of Acrocomia aculeata plants, 25 plants were measured and georeferenced in the field using a digital optical hypsometer at the experimental farm of the Federal University of Lavras. Then, a drone flight was carried out over the study area to collect images and after the processing, the Digital Surface Models (DSM) and Digital Terrain Models (DTM) were obtained, which were used to estimate the height of the macaúba plants. This data was then compared with field measurements to assess the accuracy of the estimates obtained via drone. Spearman's correlation was tested and the correlation values found were strong, above 0.81. The standard deviation of the field data and the estimate using drone ranged from 2.01 to 2.24 m. The estimation of the data using the quadratic equation resulted in MAE: 0.80 and   R2=0.70. This work highlights the potential of using drones to optimize macaúba cultivation and that the use of technologies can be a viable alternative for researchers and companies involved in the macaúba production chain. However, studies in a more robust database need to be generated.