ABSTRACT

Today, it can be seen that many of the future remote sensing processes for forestry will be based on point cloud processing or on elevation models (3D techniques). Ÿese required forestry data can be provided not only by both the lidar and the radar but also by photogrammetry. For example, analogous to photogrammetric spatial intersection, a stereo pair of SAR images with di¦erent o¦-nadir angles can be used to calculate the 3D coordinates for corresponding points on the image pair producing point clouds from radar imagery. Also, in SAR interferometry (InSAR), pixel-by-pixel phase di¦erence between two complex SAR images acquired from slightly di¦erent perspectives can be converted into elevation di¦erences of the terrain/object. Ÿus, both lidar and radar can provide data that can be processed in a similar way either using original points or using surface models in a raster form. From the point clouds, you can calculate digital terrain model (DTM), digital surface model (DSM), and canopy height model, normalized digital surface model (CHM/nDSM). Ÿe idea is to provide surface model (DSM) and subtract the ground elevation (DTM) from it in order to get a canopy height. Intensity, coherence (in interferometry SAR) and texture can be used to improve the estimates in 3D-based inventory.