Investigations toward the extraction of angular signatures from ALOS/Prism data sets

Thomas Schneider(1)

(1) Technische Universität München, Hochanger 13, D-85354 Freising, Germany


ALOS instrument configuration and data take schedule offer the unique opportunity to investigate different assessment strategies for natural and semi-natural vegetation types. The proposals general aim is to evaluate ALOS data for extracting spectral and angular signatures as well as canopy related structural information in a joint experiment. The presentation is introducing the concept, the basic methodology of “angular signature” extraction from multidirectional data sets and discusses the limitations. The approach explores the backscatter differences between nadir and off-nadir, respectively of corresponding off-nadir data sets to retrieve an additional parameter characterizing the surface element. The information derived from such “angular signatures” is of completely different nature compared to the well explored spectral signature and can be considered as an independent variable in a modelling process and assumed to be mainly related to canopy structure elements. Investigations on stand structure and plant architecture by angular signature evaluation are still at the beginning. The work flow presented is based on ALOS/Prism level 1b1 data set. Starting from a very precise ortho-rectification using the digital surface model derived from the same data set an angular signature is calculated from the ratio of the two off nadir looking bands. This angular information is used to classify homogeneous patches according to the anisotropy class as calculated from the angular signatures. Examples from a testsite in the hilly Alpine forelands South of Lake Starnberg are shown and the relation to land-cover, land use types is discussed on base of ground inspection data.



  Higher level                 Last modified: 07.05.06