Object-based classification of QuickBird data using ancillary information for the detection of forest types and NATURA 2
The detection of forest types and habitats is of major importance for silvicultural management as well as for the monitoring of biodiversity in the context of NATURA 2000. For these purposes, the presented study applies an object-based classification meth
- PDF / 1,613,794 Bytes
- 16 Pages / 439.37 x 666.142 pts Page_size
- 75 Downloads / 164 Views
M. Förster, B. Kleinschmit Berlin University of Technology, Department of Geoinformation Straße des 17. Juni 145 (EB 5), 10623 Berlin, Germany [email protected], [email protected]
KEYWORDS: QuickBird, fuzzy-logic, NATURA 2000 ABSTRACT: The detection of forest types and habitats is of major importance for silvicultural management as well as for the monitoring of biodiversity in the context of NATURA 2000. For these purposes, the presented study applies an object-based classification method using VHR QuickBird data at a test site in the pre-alpine area of Bavaria (southern Germany). Additional geo-data and derived parameters, such as altitude, aspect, slope, or soil type, are combined with information about forest development and integrated into the classification using a fuzzy knowledgebase. Natural site conditions and silvicultural site conditions are considered in this rule-base. The results of the presented approach show higher classification accuracy for the classification of forest types using ancillary information than can be reached without additional data. Moreover, for forest types with very distinctly defined ecological niches (e. g. alluvial types of forest), a better characterisation and integration of rules is possible than for habitats with very wide ecological niches. Hence, classification accuracies are significantly higher when these rules are applied. In a second step NATURA 2000 habitat types and selected habitat qualities are derived from the classified forest types. However, the share of habitat qualities varies with an altering scale. This difficulty should be addressed in further research of NATURA 2000 monitoring.
276
M. Förster, B. Kleinschmit
1 Introduction With the development of a standardised and pan-European geo-datainfrastructure (Craglia et al. 2005), remote-sensing applications which integrate GIS information will become increasingly important. Therefore, various studies of combining additional data and knowledge into classification processes (Maselli et al. 1995; Stolz 1998) were undertaken. However, the integration of additional geo-data into very high spatial resolution (VHSR) imagery remains a challenging task. Fig. 1 shows an exemplary overview of multi-scale segmentation for different forest scales with the corresponding levels of ancillary data and knowledge. Additional information can be differentiated into two categories (see Fig. 1). Firstly, spatially explicit knowledge is available. For forestry applications a broad range of this kind of data sources can be used, namely the simulation of geo-data (e.g. Disney et al. 2006; Verbeke et al. 2005), the usage of altitude information, especially with LIDAR techniques (e.g. Diedershagen et al. 2004), and the integration of silvicultural maps (e.g. Förster et al. 2005b) as well as soil and hydrology maps into classification procedures.
Fig. 1. Exemplary overview of multi-scale dependence of object-based information and ancillary GIS-Data and knowledge for a forestry application
Object-based classi
Data Loading...
