Figure 6 | Scientific Reports

Figure 6

From: Validation of UAV-based alfalfa biomass predictability using photogrammetry with fully automatic plot segmentation

Figure 6

High-throughput phenotyping alfalfa biomass using UAV images. (a) The UAV-based imagery consists of five bands from the multispectral camera and pixel-wise plant height from the Pix4D, which can derive multiple Vegetation Indices (VIs) such as NDRE, NDVI, and NGRDI. The images with multiple bands are displayed as RGB (b). NDVI was used to cluster pixels into three classes (c). The class corresponding to the vegetation area (red) is selected to define pixels of interest and displayed in RGB (d). The plot information segmented by GRID is displayed for the plots in the first column (e). GRID can correctly segment plots that are not applicable using georeferences. For example, plot A and B cannot be separated by red and yellow lines, or any horizontal lines between them (b).

Back to article page