Abstract

Historical archives of satellite imagery make it possible to make retrospective analysis of forest landscape change over large areas and long periods of time. Besides providing a historical record of the size and distribution of forest harvesting activities, a detailed map of historical changes can be used to improve estimates of the current forest state and can be an important basis for studying changing wildlife habitats and landscape patterns. While the methods for satellite remote sensing change detection are fairly well developed, there are few software tools available that can map changes from multiple scenes over large areas in an automatic way. One major challenge is how to make a coherent change product from a large set of images with different acquisition dates and different areas of overlap. In this paper we describe an ongoing project that aims at developing algorithms and a software system for mapping major forest changes from historical image archives. The approach is to find the intersection of all image pairs after applying necessary geometric and radiometric pre-processing and map major forest changes as objects with polygon topology. Each change object can then have associated metadata describing the relevant time interval and be used for further analysis and modeling in a GIS. System development is done with ESRI ArcGIS components and C#.net

Keywords

: Forest change detection; change mapping

Published in

Publisher: Skogsstyrelsen

Conference

ForestSat: Operational Tools in Forestry Using Remote Sensing Techniques

Authors' information