Using Near-Surface Observations for Optimizing the Timing of Overhead Image Acquisition for Applied Mapping of Woody Vegetation Species

We present a phenology-based approach for optimizing the number and timing of unmanned aerial vehicle imagery acquisition, based on a priori near-surface observations. A ground-placed camera was used for generating annual time series of spectral indices in four different East Mediterranean sites. The time series dataset represented 1852 individuals of 12 common vegetation species. Feature selection was used for identifying the optimal dates for species classification. A UAV was flown for acquiring five overhead multiband orthomosaics, based on the five optimal dates identified in the feature selection of the near-surface time series of the previous year. An object-based classification was used for species classification, and resulted in an average overall accuracy of 85% and an average Kappa coefficient of 0.82. This cost-effective approach has high potential for detailed vegetation mapping, regarding the accessibility of UAV-produced time series, compared to hyper-spectral imagery with high spatial resolution which is more expensive and involves great difficulties in implementation over large areas.