Estimation of regeneration coverage in a temperate forest by 3D segmentation using airborne laser scanning data

Nina Amiri, Wei Yao, Marco Heurich, Peter Krzystek, Andrew K. Skidmore

Research output: Journal article publicationJournal articleAcademic researchpeer-review

29 Citations (Scopus)


Forest understory and regeneration are important factors in sustainable forest management. However, understanding their spatial distribution in multilayered forests requires accurate and continuously updated field data, which are difficult and time-consuming to obtain. Therefore, cost-efficient inventory methods are required, and airborne laser scanning (ALS) is a promising tool for obtaining such information. In this study, we examine a clustering-based 3D segmentation in combination with ALS data for regeneration coverage estimation in a multilayered temperate forest. The core of our method is a two-tiered segmentation of the 3D point clouds into segments associated with regeneration trees. First, small parts of trees (super-voxels) are constructed through mean shift clustering, a nonparametric procedure for finding the local maxima of a density function. In the second step, we form a graph based on the mean shift clusters and merge them into larger segments using the normalized cut algorithm. These segments are used to obtain regeneration coverage of the target plot. Results show that, based on validation data from field inventory and terrestrial laser scanning (TLS), our approach correctly estimates up to 70% of regeneration coverage across the plots with different properties, such as tree height and tree species. The proposed method is negatively impacted by the density of the overstory because of decreasing ground point density. In addition, the estimated coverage has a strong relationship with the overstory tree species composition.
Original languageEnglish
Pages (from-to)252-262
Number of pages11
JournalInternational Journal of Applied Earth Observation and Geoinformation
Publication statusPublished - 1 Oct 2016
Externally publishedYes


  • 3D segmentation
  • Full waveform ALS
  • Mean shift
  • Regeneration
  • Understory

ASJC Scopus subject areas

  • Global and Planetary Change
  • Earth-Surface Processes
  • Computers in Earth Sciences
  • Management, Monitoring, Policy and Law


Dive into the research topics of 'Estimation of regeneration coverage in a temperate forest by 3D segmentation using airborne laser scanning data'. Together they form a unique fingerprint.

Cite this