Assessing a novel modelling approach with high resolution UAV imagery for monitoring health status in priority riparian forests

Forest Ecosystems

Table 3 Metrics computed from the DMS S.O.D.A-RGB-UAV-data at crown level

Class	Variable	Formula (“Terrain” package, Hijmans et al. 2015)	Reference
DSM
Slope	SLOPE_sd	Slope	(Horn 1981)
Topographic position index	TPI_mean and TPI_sd	where \( TPI=y-\sum \limits_{i=1}^8\frac{y_i}{8} \), with y is the altitude of a cell and y_i is the altitude of the i-th adjacent cell	(Wilson et al. 2007)
Terrain ruggedness index	TRI_mean and TRI_sd	where \( TRI=\sum \limits_{i=1}^8\frac{\left\|{y}_i-y\right\|}{8} \), with y is the altitude of a cell and y_i is the altitude of the i-th adjacent cell
Roughness	ROUGHNESS_mean and ROUGHNESS_sd	Roughness is the difference between the maximum and the minimum value of a cell and its 8 surrounding cells
Texture	dsm_GLCM_mean (window size of 3 × 3 pixels and a 45 degree shift)	“mean”, “variance”, “homogeneity”, “contrast”, “dissimilarity”, “entropy”, “second_moment”, “correlation”