Replacing sample plots forest inventory by whole stand measurements based on lidar and orthophoto

Abstract

Automatic tree counting algorithms based upon image segments form the core of accessing yield per area unit in a standing stock. The Oject Based Image Analysis specifies unique transferable attributes of the crown ‘hotspot‘, which spatial characteristics are encounter over a wide range of aerial photographs, orthophoto‘s and satellite imagery. With high quality data, including LiDAR height information, a new large scale yield estimation can take place using semi and/or full automatic procedures. The method was tested on two large test area‘s after development on a variety of data sets within the laboratory of RS and GIS at the faculty of forestry in Cracow