Refine
Year of publication
- 2012 (2)
Document Type
- Article (2)
Language
- English (2)
Has Fulltext
- yes (2)
Is part of the Bibliography
- yes (2)
Keywords
- Lidar (2)
- Forstwirtschaft (1)
- Fotogrammetrie (1)
Institute
In this paper, we present an integrated strategy to comprehensively evaluate the feature relevance of point cloud and image data for classification of trees and vehicles in urban scenes. First of all, point cloud and image data are co-registered by backprojection with available orientation parameters if necessary. After that, all data points are grid-fitted into the raster format in order to facilitate acquiring spatial context information per pixel/point. Then, various spatial-statistical and radiometric features can be extracted using a cylindrical volume neighborhood. Classification results as labeled pixels can be acquired from the classifier, and after appropriate refinements we obtain the objects of trees and vehicles. Compared to other methods which have assessed the classification and relevance simultaneously using a single classifier, we first introduce AdaBoost classifier combined with contribution ratio to provide both classification results and measures of feature relevance, and then utilize Random Forest classifier to evaluate and compare the feature relevance from a more independent viewpoint. In order to confirm the accuracy and reliability of classification and feature relevance results, we consider not only characteristics of the classifiers itself, but also errors of data co-registration and alterable parameters. We apply the procedure to two different datasets. In the dataset requiring co-registration a-priori, the AdaBoost classifier even achieves a great accuracy of 96.99% for trees and 83.45% for vehicles. The quantitative results of feature relevance assessment highlight the most important features for classification of tree covers and vehicles, such as NDVI, LiDAR intensity, planarity and entropy. By comparative analysis of the two independent approaches, the reliable and consistent feature selection for classification of trees and vehicles from LiDAR and image data could be validated and achieved, being unrelated to the classifiers.
In forest ecology, a snag refers to a standing, partly or completely dead tree, often missing a top or most of the smaller branches. The accurate estimation of live and dead biomass in forested ecosystems is important for studies of carbon dynamics, biodiversity, and forest management. Therefore, an understanding of its availability and spatial distribution is required. So far, LiDAR remote sensing has been successfully used to assess live trees and their biomass, but studies focusing on dead trees are rare. The paper develops a methodology for retrieving individual dead trees in a mixed mountain forest using features that are derived from small-footprint airborne full waveform LIDAR data. First, 3D coordinates of the laser beam reflections, the pulse intensity and width are extracted by waveform decomposition. Secondly, 3D single trees are detected by an integrated approach, which delineates both dominate tree crowns and understory small trees in the canopy height model (CHM) using the watershed algorithm followed by applying normalized cuts segmentation to merged watershed areas. Thus, single trees can be obtained as 3D point segments associated with waveform-specific features per point. Furthermore, the tree segments are delivered to feature definition process to derive geometric and reflectional features at single tree level, e.g. volume and maximal diameter of crown, mean intensity, gap fraction, etc. Finally, the spanned feature space for the tree segments is forwarded to a binary classifier using support vector machine (SVM) in order to discriminate dead trees from the living ones. The methodology is applied to datasets that have been captured with the Riegl LMSQ560 laser scanner at a point density of 25 points/m2 in the Bavarian Forest National Park, Germany, respectively under leaf-on and leaf-off conditions for Norway spruces, European beeches and Sycamore maples. The classification experiments lead in the best case to an overall accuracy of 73% in a leaf-on situation and 71% in a leaf-off situation, if we assess the classification results using 5-fold cross-validation method with the help of reference data acquired by the field surveying.