Single tree detection and modeling using airborne laser scanner data [Elektronische Ressource] / by Sandeep Gupta

Single Tree Detection and Modeling Using Airborne Laser Scanner Data Thesis submitted in partial fulfillment of the requirements of the degree Doctor rer. nat. of the Faculty of Forest and Environmental Sciences, Albert-Ludwigs-Universität Freiburg im Breisgau, Germany By Sandeep Gupta Freiburg im Breisgau, Germany 2010 Dean: Prof. Dr. Dr. h.c. Gero Becker Supervisor: Prof. Dr. Barbara Koch nd2 Reviewer: Prof. Dr. Dr. h. c. Dieter Rudolf Pelz Disputation: 07 December 2010 ABSTRACT The increasing demand for single tree based information, as a basis to improve forest management practices, is the motivation behind the development of single tree detection and modeling techniques that processes airborne laser scanning (ALS) data. Most tree detection algorithms are raster based while fewer employ vector-based approaches. Among the vector-based methodologies, little attention has been paid on the potential of clustering mechanisms. Clustering since long has been considered as a pattern classification technique that groups similar patterns using a chosen distance measure. The k-means has been considered as a staple of clustering algorithms and has been found useful in tree detection.