Mapping Forest Regeneration from Terrestrial Laser Scans

Abstract

Location, spread, abundance and density of forest regeneration are key factors in understanding forest dynamics as well as in operational management of uneven-aged stands. Simulation of forest growth, silviculture and planning of skid road networks require accurate and objective methods for locating forest regeneration. Terrestrial laser scanning has high potential for tree mapping, however, the development of automatic processing methods has been focused on mature trees so far. This study introduces an automatic procedure to locate individual trees with 3-6 meter height from terrestrial laser scanner data. The method has been validated on three sample quadrates representing different stand structures and it succeeded in detecting 79-90% of trees extracted manually from the point cloud. Out of the investigated stand features, stem density had the strongest impact on the performance, while branching intensity slightly affected the detection rate. The results highlight that terrestrial laser scanning has the ability for the quantitative evaluation of regeneration, providing a prospective tool for surveying forests of contiguous cover.

Detecting Tree Stems from Volumetric TLS Data in Forest Environments with Rich Understory

“…Founded on high-resolution data of a single tree, their approach is not suitable to manage larger occlusions within datasets from technically challenging forest conditions. The scale and structure of the approach from Brolly et al [32] is more similar to ours, but differs in the techniques applied. Compared to our series of very strict morphological probing, the authors used a single, relatively tolerant filter.…”

“…It includes logical processing chains by making use of the ordered grid structure, which is characterized by a discrete topology of the data and the principles of connectivity and adjacency. While many studies on tree stem detection applied classification and shape fitting algorithms to the point cloud [4,20,26,33,35], only a few studies follow morphological approaches on voxel grids [28,29,32]. These studies give an indication that morphological methods have good potential to extract stems from layered forest structures and should be observed further.…”

“…For this reason, many approaches for stem detection in forest environments concentrated on free-standing mature trees. Only a few studies exist that tried to detect stems that were occluded by understory layers or regeneration trees, as for example Brolly et al [32] and Xia et al [33]. While both studies are innovative in their field, they dealt with simplified scenarios.…”

“…While both studies are innovative in their field, they dealt with simplified scenarios. Brolly et al [32] worked with homogenous regeneration patches only, excluding mature trees. It also remains unclear if they used data from foliated or non-foliated trees.…”

Extraction of Urban Trees From Integrated Airborne Based Digital Image and Lidar Point Cloud Datasets - Initial Results

“…The capability of LiDAR to pass through vegetation has attracted remarkable concern from the field of natural resource management (Gaulton, et al, 2010, Hudak, et al, 2009, Liang, et al, 2007. From a forest management stand-point, LiDAR has been used to define information about trees (Coops et al, 2007, Brolly, et al, 2013, Lang, et al, 2006, measure carbon stocks (Patenaude et al, 2004), compute fuel quantity (Seielstad and Queen, 2003) and create habitat models (Vierling, et al, 2008), develop forest inventories (Zhang, C., 2010, Woods, et al, 2008. Even though considerable research has been carried out regarding LiDAR applications in forestry, its usage in the study of urban trees has been limited.…”