University of Twente Student Theses

Login

Detecting changes in trees using multi-temporal airborne lidar point clouds

Xiao, Wen (2012) Detecting changes in trees using multi-temporal airborne lidar point clouds.

[img] PDF
3MB
Abstract:Changes in vegetation are of great interest since they play crucial roles in ecosystem monitoring where remotely sensed data has been proven extremely profitable. Digital change detection approach has been widely utilised in conventional remote sensing technologies. As a relatively new technology, light detection and ranging (lidar) provides a promising way of change detection of vegetation in three dimensions (3D) because the laser beam will penetrate through the foliage generating point clouds with highly accurate 3D coordinates. This research proposes a method for vegetation change detection in 3D with high level of automation. Three epoch datasets are classified into several predefined classes including high vegetation (trees). A connected components algorithm is applied to group the points of a tree together because the point clouds are unstructured. The attributes of components are used to discriminate tree components from other since a few non-tree points are misclassified. Points from neighbour trees might be clustered together, so a local maxima algorithm is implemented to distinguish single tree components with multiple tree components. After that, the parameters of trees are derived through two independent ways: point based method which refers to 3D alpha shapes and convex hull; model based method which utilises the Pollock tree model for single trees. Then the changes can be detected by comparing the parameters of corresponding tree components which are found by a tree to tree matching algorithm. The comparison of these two methods illustrates the consistency and stability of the parameters. The detected changes clearly show the growth and pruning of trees. The results are visualized by point cloud mapper (PCM) and statistically analysed. Keywords: change detection, high vegetation, 3D modelling, airborne lidar, point cloud
Item Type:Essay (Master)
Faculty:ITC: Faculty of Geo-information Science and Earth Observation
Programme:Geoinformation Science and Earth Observation MSc (75014)
Link to this item:https://purl.utwente.nl/essays/93606
Export this item as:BibTeX
EndNote
HTML Citation
Reference Manager

 

Repository Staff Only: item control page