New possibilities for analysing forest structure by remote sensing
Abstract
The Austrian forestry faces new challenges with regard to the foundation and management of site-adapted forests. In this context, planning- and decision-processes take place, where large-scale geodata that are assessed in a standardized way and are available on short notice, become more and more important. Forest remote sensing has already proved its suitability for fulfilling practical needs for many years. The potential of modern remote sensing data however is in use only to a small degree. This is especially true for the automated analysis of data from digital airborne cameras. In airborne imagery the appearance of an object varies considerably in accordance with various illumination situations (back light, side light, …), which makes automated image analysis difficult. Although it is known that there is a correlation between these effects and the structural properties of the displayed objects, this fact is rarely used for classification because suitable algorithms are missing. The project aims at making use of the described phenomenon for an improved classification of forest parameters instead of considering such effects as disturbing artefacts and eliminating them as usually done so far. The project is based on own research results and focuses on developing an automated classification procedure. Digital aerial images as provided by standard aerial surveys are used. Das Projekt wird durch einen Preis vergeben durch die Stiftung "120 Jahre Universität für Bodenkultur" unterstützt.
keywords digital airborne camera digital photogrammetry Remote Sensing forest structure
Publikationen
Project staff
Clement Atzberger
Univ.Prof. Dr.rer.nat. Clement Atzberger
clement.atzberger@boku.ac.at
Tel: +43 1 47654-85711
Project Leader
27.04.2017 - 30.06.2017