A Novel Vegetation Index for Coffee Ripeness Monitoring Using Aerial Imagery

Coffee ripeness monitoring is a key indicator for defining the moment of starting harvest, especially because coffee quality related to fruit degree. The most used method define start harvesting by visual inspection, which time-consuming, labor-intensive, and does not provide information on entire area. There lack new techniques or alternative methodologies faster measurements that can support harvest planning. Based that, this study aimed at developing vegetation index (VI) using aerial imagery. For this, an experiment was set up in five arabica fields Minas Gerais State, Brazil. During stage, four flights were carried out acquire spectral crop canopy two quadcopters, one equipped with five-band multispectral camera another RGB (Red, Green, Blue) camera. Prior flights, manual counts percentage unripe fruits irregular sampling grids each day validation purposes. After image acquisition, (CRI) other VIs obtained. CRI developed combining reflectance from red band ground-based target placed effectiveness compared under different analyses traditional VIs. showed higher sensitivity discriminate plants ready not-ready all fields. Furthermore, highest R2 lowest RMSE values estimating also presented (R2: 0.70; 12.42%), whereas ranging 0.22 0.67 13.28 16.50, respectively. Finally, demonstrated time-consuming fieldwork be replaced methodology based