Urban Tree Classification Using Full-Waveform Airborne Laser Scanning

Koma, Zsófia; Koenig, Kristina; Höfle, Bernhard

doi:10.5194/isprsannals-iii-3-185-2016

Cited by 7 publications

(4 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The collinearity analysis in combination with the RFE procedure allowed us to reduce the number of lidar metrics in the classification at each hierarchical level. A similar approach to metric reduction has been used in other lidar‐based vegetation classifications (Millard and Richardson, 2013; Alexander et al, 2016; Koma et al, 2016). In our study, only five lidar metrics related to vegetation structure where needed to separate different land cover types and habitats within this wetland.…”

Section: Discussionmentioning

confidence: 99%

Classifying wetland‐related land cover types and habitats using fine‐scale lidar metrics derived from country‐wide Airborne Laser Scanning

Koma

Seijmonsbergen

Kissling

2020

Remote Sens Ecol Conserv

View full text Add to dashboard Cite

Mapping 3D vegetation structure in wetlands is important for conservation and monitoring. Openly accessible country‐wide Airborne Laser Scanning (ALS) data—using light detection and ranging (lidar) technology—are increasingly becoming available and allow us to quantify 3D vegetation structures at fine resolution and across broad spatial extents. Here, we develop a new, open‐source workflow for classifying wetland‐related land cover types and habitats using fine‐scale lidar metrics derived from country‐wide ALS data. We developed a case study in the Netherlands with a workflow consisting of four routines: (1) pre‐processing of ALS data, (2) calculation of lidar metrics (i.e. 31 features representing cover, 3D shape, vertical variability, horizontal variability and height of vegetation as well as microtopography), (3) assessing feature importance of lidar metrics for classifying wetland habitats, and (4) applying a Random Forest algorithm for mapping and prediction. We used an expert‐based vegetation map for annotation and generated 100, 500 and 1000 annotation points for each class. Using a three‐level hierarchical approach, we differentiated at level 1 planar surfaces (e.g. roads and agricultural fields) from wetland vegetation with 82% mean overall accuracy, using predominantly height and horizontal variability metrics. At level 2, we classified wetland vegetation into four land cover types (forest, grassland, reedbeds, shrubs) with 71% mean overall accuracy, using lidar metrics related to vegetation height and horizontal and vertical variability. At level 3, we differentiated two types of land reed as well as water reed with 78% mean overall accuracy, using predominantly vertical variability metrics. Our results demonstrate that lidar metrics (related to vegetation height, cover, vertical and horizontal variability) derived from country‐wide ALS data can differentiate land cover types and habitats within wetlands at high resolution. Given appropriate annotation data, our workflow can be up‐scaled to a country‐wide extent to allow the comprehensive mapping and monitoring of wetlands at national scales.

show abstract

Section: Discussionmentioning

confidence: 99%

Classifying wetland‐related land cover types and habitats using fine‐scale lidar metrics derived from country‐wide Airborne Laser Scanning

Koma

Seijmonsbergen

Kissling

2020

Remote Sens Ecol Conserv

View full text Add to dashboard Cite

show abstract

“…However, the authors warn that the algorithms are over-adapted for the specific forest type. Studies using data derived from LiDAR or Unmanned Aerial Vehicles (UAV) cite slightly higher accuracies ranging from 80% to 90% [9,18,[66][67][68][69], while recent deep learning developments reached accuracies of >90% [4,70,71].…”

Section: Discussionmentioning

confidence: 99%

A Method for Tree Detection Based on Similarity with Geometric Shapes of 3D Geospatial Data

Stupariu

Pleșoianu

Pătru-Stupariu

et al. 2020

IJGI

View full text Add to dashboard Cite

This paper presents an approach to detecting patterns in a three-dimensional context, emphasizing the role played by the local geometry of the surface model. The core of the associated algorithm is represented by the cosine similarity computed to sub-matrices of regularly gridded digital surface/canopy models. We developed an accompanying software instrument compatible with a GIS environment which allows, as inputs, locations in the surface/canopy model based on field data, pre-defined geometric shapes, or their combination. We exemplified the approach for a study case dealing with the locations of scattered trees and shrubs previously identified in the field in two study sites. We found that the variation in the pairwise similarities between the trees is better explained by the computation of slopes. Furthermore, we considered a pre-defined shape, the Mexican Hat wavelet. Its geometry is controlled by a single number, for which we found ranges of best fit between the shapes and the actual trees. Finally, a suitable combination of parameters made it possible to determine the potential locations of scattered trees. The accuracy of detection was equal to 77.9% and 89.5% in the two study sites considered. Moreover, a visual check based on orthophotomaps confirmed the reliability of the outcomes.

show abstract

“…The points in the first zone were manually classified when classifying landscape vegetation. The automated process was set to exclude the lowest zone because of interference of urban reflecting objects (47,48). Buildings: Buildings and houses enter the streetscape peripherally.…”

Section: Methodsmentioning

confidence: 99%

Measuring Streetscape Features with High-Density Aerial Light Detection and Ranging

Golombek

Marshall

2020

Transportation Research Record

View full text Add to dashboard Cite

This study investigates the feasibility of extracting streetscape features from high-density United States Geological Survey (USGS) quality level 1 (QL1) light detection and ranging (LiDAR) and quantifying the features into three-dimensional (3D) volumetric pixel (voxel) zones. As the USGS embarks on a national LiDAR database with the goal of collecting LiDAR across the continuous U.S.A., the USGS primarily requires QL2 or QL1 as a collection standard. The authors’ previous study thoroughly investigated the limits of extracting streetscape features with QL2 data, which was primarily limited to buildings and street trees. Recent studies published by other researchers that utilize advanced digital mapping techniques for streetscape measuring acknowledge that most features outside of buildings and street trees are too small to detect. QL1 data, however, is four times denser than QL2 data. This study divides streetscapes into Thiessen proximal polygons, sets voxel parameters, classifies QL1 LiDAR point cloud data, and computes quantitative statistics where classified point cloud data intersects voxels within the streetscape polygons. It demonstrates how most other common streetscape features are detectable in a standard urban QL1 dataset. In addition to street trees and buildings, one can also legitimately extract and statistically quantify walls, fences, landscape vegetation, light posts, traffic lights, power poles, power lines, street signs, and miscellaneous street furniture. Furthermore, as these features are processed into their appropriate voxel height zones, this study introduces a new methodology for obtaining comprehensive tabular descriptive statistics describing how streetscape features are truly represented in 3D.

show abstract

Urban Tree Classification Using Full-Waveform Airborne Laser Scanning

Cited by 7 publications

References 17 publications

Classifying wetland‐related land cover types and habitats using fine‐scale lidar metrics derived from country‐wide Airborne Laser Scanning

Classifying wetland‐related land cover types and habitats using fine‐scale lidar metrics derived from country‐wide Airborne Laser Scanning

A Method for Tree Detection Based on Similarity with Geometric Shapes of 3D Geospatial Data

Measuring Streetscape Features with High-Density Aerial Light Detection and Ranging

Contact Info

Product

Resources

About