General Tools - Digital Elevation Model Fusion - Point Cloud DEM Fusion

<< Click to Display Table of Contents >>

Navigation:  Help Content > General Tools > DEM Fusion >

General Tools - Digital Elevation Model Fusion - Point Cloud DEM Fusion

Previous pageReturn to chapter overviewNext page

Purpose

 

Two existing DEMs in Point Cloud form (Shapefile), previously generated with SARscape or OPTICALscape, can be combined into a single raster product.

 

Technical Note

 

The process initially exploits the information provided by the respective DEM/DSM generation modules provided in OPTICALscape and SARscape to apply both a knowledge-based and an accuracy-based sample selection. The choice strategies vary depending on the type of process:

 

Optical-Optical

In the case of Optical data, the information contained in the shapefile consists in the Feature Type (grid, point, edge; for details refer to the OPTICALscape documentation), matching cross-correlation and reliability index. Exploiting the knowledge about stereo-optical matching, an edge or point feature will be always chosen rather than a grid point. The choice between identical features is driven by cross-correlation and reliability index.

 

Optical-SAR

The knowledge based part is exploited in this process as well. An edge feature found in optical images will be chosen against a SAR produced data point. On the opposite, if the optical data point represents a grid point, the SAR data will be preferred. In the case of points equally important, a comparison between the SAR precision and the Matching cross-correlation will be performed.

 

SAR-SAR

The choice will be solely data-driven, the excluded point will always be the one showing the worst precision.

 

The final step will be performed exploiting a modified 2D Sheppard local interpolator using Thin Plate Radial Basis Functions, the local RBF support will be automatically set. The approach is inspired by the one proposed by  Lazzaro D. et. al., the RBF function is given in Numerical Recipes. Two parameters have to be set by the user, namely the number of nearby points on which to fit the RBF function and the number of nodes defining the radius of influence of the computed interpolant. These two values are usually set to 13 and 10 by default respectively.

 

Two additional steps can be performed, namely a Mean Filtering and a Relaxation interpolation. The former in order to smooth eventual spikes and the latter to provide a continuous surface without holes.

 

It is compulsory to provide point clouds produced in the same cartographic system.

 

Input Files

 

DEM Point Cloud File 1

File name of the higher resolution input point cloud. Point cloud in shapefile format (file size lower that 2 GB) and LAS format (file size higher that 2GB) created in SARscape or OpticalScape are supported. For Optical-SAR processing, this file must be the Optical input. This file is mandatory.

 

DEM Point Cloud File 2

File name of the lower resolution input point cloud. Point cloud in shapefile format (file size lower that 2 GB) and LAS format (file size higher that 2GB) created in SARscape or OpticalScape are supported. For Optical-SAR processing, this file must be the Optical input. This file is mandatory.
 

Important: - It is compulsory to provide point clouds produced in the same cartographic system.

- Only LAS files produced with sarmap software are supported.

 

DEM/Cartographic System

 

Output Projection

The following parameters are compulsory to define the Cartographic System:

 

State

Definition of the country or general projection systems.

 

Hemisphere

Definition of the hemisphere. This is chosen according to the selected State and Projection.

 

Projection

Definition of the projection system of the selected State. In case that a general projection system is selected in State, the Projection is automatically set.

 

Zone

Definition of the zone. This is chosen according to the selected State and Projection.

 

Ellipsoid

Definition of the ellipsoid. This is chosen according to the selected State and Projection.

 

Datum Shift Parameters

Definition of the datum shift parameters. These are chosen according to the selected State and Projection.

 

Note that the inputs have to be in the same cartographic system as the output.

 

Parameters - Principal Parameters

 

Dem Data Type1

This flag must be set to define the input type of the high resolution DEM, the choice is given between:

 

SAR, if the corresponding input is an interferometric DEM (necessarily generated with SARscape).

Optical, if the corresponding input is a stereo DEM (necessarily generated with OPTICALscape).

 

Dem Data Type2

This flag must be set to define the input type of the low resolution DEM, the choice is given between:

 

SAR, if the corresponding input is an interferometric DEM (necessarily generated with SARscape).

Optical, if the corresponding input is a stereo DEM (necessarily generated with OPTICALscape).

 

Number Of Coefficient Points

Number of nearby points for the fitting on each node.

 

Number Of Nodes

Number of nodes defining the radius of influence of the interpolant.

 

Grid Size

The pixel spacing of the output DEM. This value is mandatory.

 

Relax Interpolation

Set this flag if a final relax interpolation step has to be executed in order to fill eventual holes..

 

Mean Filtering

Set this flag if a mean filtering on a regular grid has to be executed.

 

Mean Window Size

The size of the window for mean filtering. This value is set as 3x3 by default.

 

Interpolator

The choice is given between:

Radial Basis Function (RBF).

Inverse distance weighted Least Squares (LS)

 

Point Cloud Registration

Set this flag if you wish to register the point clouds before filtering and interpolation. Set this flag only if the input point clouds show a shift between them.

 

Registration Iterations

The number of iterations for the registration step.

 

Point Cloud Filtering

Set this flag if the cloud shows evident noise aberrations that you wish to mitigate. The output point cloud will include less points than the input, single isolated points will be eliminated. The Filter radius and Points in radius parameters must be set in accordance with the density of the input point clouds (how many points should be in that radius).

 

Filter Radius (m)

The radius for the point cloud filter.

 

Points in radius

The number of points to consider for the given radius in the point cloud filter.

 

Interpolation Distance Constraint

Set this flag if you wish to constrain the interpolation to a certain distance from the point observations. This will mitigate the edge effect of the interpolator for sparse clouds.

 

Interpolation Distance Multiplier

Set this flag if you wish to constrain the interpolation to a certain distance from the point observations. This will mitigate the edge effect of the interpolator for sparse clouds.

 

Parameters - Global

 

It brings to the general section of the Preferences parameters. Any modified value will be used and stored for further processing sessions.

 

Parameters - Other Parameters

 

It brings to the general section of the Preferences parameters. Any modified value will be used and stored for further processing sessions.

 

Output Files

 

Output File

File name of the output combined Digital Elevation Model. This file is mandatory.

 

_PC_dem

Fusion resulting Digital Elevation Model with the corresponding header files (.sml, .hdr).

 

General Functions

 

Exec

The processing step is executed.

 

Store Batch        

The processing step is stored in the batch list. The Batch Browser button allows to load the batch processing list.

 

Close        

The window will be closed.

 

Help

Specific help document section.

 

 

Specific Function(s)

 

None.

 

See Also

 

Task, SARscapeBatch object, SARscapeBatch script example

 

References

 

Lazzaro D., Montefusco L.B., Radial Basis Functions for the multivariate interpolation of large scattered data sets, Journal of Computational and Applied Mathematics, 140, pages 521-536, 2002.

Numerical Recipes. The Art of Scientific Computing, 3rd Edition, 2007.