New methods of remote sensing have strongly influenced forest applications in the fields of conservation, monitoring and assessment of forest ecosystems. From assessing forest ecosystem health, over protecting and preserving biodiversity, to monitoring single trees and entire forests—gaining accurate information on the status and distribution of forest structures over various time scales is vital. Active and passive sensors such as LiDAR and optical sensors capture the three-dimensional forest structure in geometric and multispectral detail. These instruments operate with high point density, and are available as aerial, terrestrial and mobile devices, allowing for new approaches and applications. The miniaturization of sensors and platforms in combination with increased measurement speed opens up new application possibilities with a clear tendency toward drone-based projects. New deep learning methods outperform classic machine learning techniques and lead to new methodological approaches with a significant increase in accuracy.

In this special issue, we present three articles that address forestry applications on the one hand and a new sensor system on the other. The paper of Hell et al. compares two deep learning-based methods that classify tree species and standing dead trees from laser point clouds. The second paper of Kedra et al. presents a method that analyzes the tree structure of spruce trees (Picea abies) from smartphone images. Finally, the report by Jenal et al. highlights a modular sensor carrier system suitable as a remote-sensing platform for vegetation monitoring.