Estimation of Urban Green Volume Based on Single-Pulse LiDAR Data

Estimating urban green volume is getting more and more important within the frame of an ecologically orientated city planning and environmentally sustainable development. The first and the last pulse of airborne Light Detection and Ranging (LiDAR) data provide the basis for the estimation of green volume, but these optimal data are not always available, particularly for urban areas. That is why this paper deals with the question whether LiDAR data (last pulse only) that have not been taken during the vegetation period allow a sufficient estimation of the green volume. This paper sets up on previous results where LiDAR data have been compared to photogrammetrically determined vegetation height measurements. The subtraction of the laser-based Digital Terrain Model and Digital Surface Model in vegetated areas leads to a vast underestimation of green volume of up to 85%, which is mainly due to the standing deciduous trees with an underestimation of 90%. Starting from the existence of different laser response characteristics of various vegetation types, the relative point density and the normalized height of classified nonground points were analyzed in depth. The results show a good separation of different vegetation types. Furthermore, a pragmatic approach of reconstruction of the underestimated vegetation (mainly deciduous trees) is carried out by generating cylinders for the classified nonground points to compensate the volume loss. The point density of nonground points and the normalized height of the laser responses were used to regulate the adaptive cylinder construction based on fuzzy logic techniques. Using reference data, the accuracy could be estimated. In spite of the suboptimal LiDAR data, this paper leads to a sufficiently exact and efficient estimation of green volume compared to the costly conventional methods like field investigations. The method makes a contribution in the field of data improvement and is applicable to similar LiDAR data of other areas.

[1]  K. Kraus,et al.  Determination of terrain models in wooded areas with airborne laser scanner data , 1998 .

[2]  Werner Pillmann,et al.  Urban Green Space Management Information – Processing and Use of Remote Sensing Images and Scanner Data , 2002 .

[3]  Chengcui Zhang,et al.  A progressive morphological filter for removing nonground measurements from airborne LIDAR data , 2003, IEEE Trans. Geosci. Remote. Sens..

[4]  Roberto Manduchi,et al.  Supervised Parametric Classification of Aerial LiDAR Data , 2004, 2004 Conference on Computer Vision and Pattern Recognition Workshop.

[5]  P. Litkey,et al.  Algorithms and methods of airborne laser-scanning for forest measurements , 2004 .

[6]  Thomas Blaschke,et al.  A “ Green Index ” Incorporating Remote Sensing and Citizen ’ S Perception of Green Space , 2005 .

[7]  Reconstruction of Urban Vegetation Based on Laser Scanner Data at Leaf-Off Aerial Flight Times – First Results , 2005 .

[8]  W. Wagner,et al.  3 D VEGETATION MAPPING AND CLASSIFICATION USING FULL-WAVEFORM LASER SCANNING , 2006 .

[9]  K. Hannemann,et al.  Monitoring the urban development with integrated system from RS observation and GIS information , 2007, 2007 Urban Remote Sensing Joint Event.

[10]  J. Reitberger,et al.  COMBINED TREE SEGMENTATION AND STEM DETECTION USING FULL WAVEFORM LIDAR DATA , 2007 .

[11]  T. Blaschke,et al.  Quantifying and Qualifying Urban Green by Integrating Remote Sensing, GIS, and Social Science Method , 2008 .

[12]  Lotfi A. Zadeh,et al.  Fuzzy Logic , 2009, Encyclopedia of Complexity and Systems Science.