Terrain Flattening Operator

When land cover classification is applied to terrain that is not flat, inaccurate classification result is produced. This is because that terrain variations affect not only the position of a target on the Earth's surface, but also the brightness of the radar return. Without treatment, the radiometric biases caused by terrain variations are introduced into the coherency and covariance matrices. It is often seen that the classification result mimic the radiometry rather than the actual land cover. This operator removes the radiometric variability associated with topography using the Radiometric Terrain Correction algorithm proposed by Small [1] while leaving the radiometric variability associated with land cover.

In the Radiometric Terrain Correction algorithm [1], the radiometric effect is simulated using a digital elevation model (DEM) of the imaged area. It is therefore required that the DEM resolution must be higher than the image resolution. In case that the DEM resolution is lower than the image resolution, users have two options:

Oversample the DEM to higher resolution, which generally will lead to longer the processing time and occasional artefacts in the image;
Multilook the source image to lower resolution, which will reduce the processing time and produce generally better image.

Input and Output

The input to this operator should be calibrated beta0. For polarimetric SAR product, it should be in T3, C3, C2 matrix format.
The output of this operator is terrain flattened gamma0. For polarimetric SAR product, the output is terrain flattened coherency or covariance matrices in T3, C3 or C2 format.

Parameters Used

The following processing parameters are used for this operator:

Source Bands: source product bands. User can select one or more bands for terrain flattening. If no band is selected, by default all bands are selected for terrain flattening.
Digital Elevation Model: available digital elevation models used in computing local illuminated area.
DEM resampling Method: resampling method used in getting elevation from DEM.
External DEM: User specified external DEM file. Currently only DEM in Geotiff format with geographic coordinates (P_lat, P_lon, P_h) referred to global geodetic ellipsoid reference WGS84 (and height in meters) is supported.
DEM No Data Value: No Data Value for user specified external DEM.
External DEM Apply EGM: checkbox if selected, Earth Gravitional Model will be applied to the user specified external DEM.
Additional Overlap Percentage: To perform terrain flattening to a given tile, pixels from adjacent tiles are generally needed due to the topography in the image area. The overlap percentage is automatically computed using the DEM. However, if the computed overlap is not enough, then tiling effect can be observed in the terrain flattened image. In this case, user can increase the Additional Overlap Percentage to increase the overlap between adjacent tiles.
Oversampling Multiple: The Terrain Flattening algorithm requires that the DEM resolution is higher than the image resolution. Therefore, the DEM is generally oversampled. The default oversampling factor is automatically computed based on the DEM resolution and image pixel spacings. The final oversampling factor = (default oversampling factor) x (Oversampling Multiple). The default value for Oversampling Multiple is 1.0. However, if the oversampling factor is not large enough, then artefacts can be observed in the terrain flattened image. In this case user can increase the oversampling factor by setting Oversampling Multiple to a larger value.

Reference:

[1] David Small, "Flattening Gamma: Radiometric Terrain Correction for SAR imagery", IEEE Transaction on Geoscience and Remote Sensing, Vol. 48, No. 8, August 2011