Dataset of digital cover photography (DCP) images of beech (Fagus sylvatica) canopies

Published: 10 June 2022| Version 1 | DOI: 10.17632/w6ptkf48jr.1
Francesco Chianucci


The dataset contains 315 cover photography (DCP) images acquired in seven 0.25 ha pure beech (Fagus sylvatica L.) forest stands sampled in central Italy (Chianucci 2020). The stands showed differences in canopy structure, resulting from different silvicultural regimes, with an effective Le estimated from reference terrestrial laser scanning (TLS) data ranging between 1.3 and 5.3. In each stand, 9 measurements were acquired on a grid using TLS. Five DCP images per scan point were collected, with the first image acquired in the same TLS point location, and the other images acquired at four cardinal directions, within 5 m from each scan point. Images were processed using the coveR package of R (Chianucci et al. 2022), which is available at: The folder 'results' showed the classifed images, which were obtained from the process_image.R code