Key Parameters


Scenario

Scenarios - optimize parameters for certain applications

  1. Default - for unstructured, close range or UAV datasets (or if the scenarios below do not apply)
  2. Nadir - for line-structured nadir camera systems in airborne applications
  3. Oblique - for line-structured oblique camera systems with 1 Nadir and 4 Oblique cameras

See also the Scenario page for details.

GSD - Ground Sample Distance

Metric raster cell size on ground level.

CLI
-gsd <value>

Default value(s): determined automatically unless manually set

The GSD determines the resolution of the tiled outputs, e.g. DSM or True Ortho. Unless manually set it will be determined automatically during the project's analysis stage. Note that the automatically determined effective GSD may differ significantly from the actual input image's pixel size on ground level, depending very much on the specified scenario.

Automatic GSD multiplicator for low overlap scenarios

When processing aerial scenario with less than 80% overlap specified in flight direction, the GSD will be doubled automatically (quarter resolution) in order to guarantee for completeness of the results.

Area

Set the project's area/bounding box in 2D/3D. Refers to the domain in object space to be processed within the project. For further details regarding functionality, refer to this page.

CLI
-area2d <xmin> <xmax> <ymin> <ymax>
-area3d <xmin> <xmax> <ymin> <ymax> <zmin> <zmax>
-area_shape <path to shape file (*.shp)>
-no_area_shape

Default value(s): determined automatically during project's analysis stage if not manually specified.

The generated cloud and DSM tiles will be aligned to the upper left corner of the bounding box of the project's processing area. Specifying a Region Of Interest within a Project (whether it captures only a subset for testing purposes, or it is limited to the actual objects from the real world intended to be captured) can potentially speed up the processing dramatically. This is because the user has the option to impose an automatic image preselection for processing, in addition to the fact that the results are filtered based on area; moreover, a masking of the image content based on Project ROI occurs, such that only those pixels, that the ROI is projected on, will be matched to generate 3D points. 

Waterbody shape file

Optionally specify a shape file containing waterbody polygons.

CLI
-waterbody <path to shape file (*.shp)>

Default value(s): none

For further information, see the articles on shape file specification/format and the full tutorial on the usage of waterbody shape files to enhance DSMs, True Orthos and Meshes.

Project Coordinate System

Specify the Project Coordinate System in WKT format (Well-Known-Text). 

Important

The information given in the WKT string defines the coordinate system your input data is given in (e.g. shape files, project coordinate system) and will also be attached to your output data (e.g. TIFF headers, point cloud headers). Specifying a different coordinate system than the input data is actually given in will not transform your output data into that system.

For more information, refer to Well-known-text specification.

Color Adjustment

Specify if you want to calculate the color adjustment parameters. They are calculated by default within the new version during Analysis. The user has to decide only if the color adjustment parameters will be applied on the output TrueOrtho and Meshes.

-color_adjustment
-no_color_adjustment
CLI
-coordinate_system <path>
-no_coordinate_system 

Tile size

Set the project's metric tile size in 2D/3D.

CLI
-tilesize <value>

Default values(s): depending on scenario and GSD, see table below.

In order to process large scale datasets, a tiling scheme will be employed to subdivide the project's area into equally sized quadratic/cubic tiles. Unless manually specified, the tile size will be automatically determined during the project's analysis stage, depending on the specified scenario and GSD. The output tiles will be aligned to the lower left corner of the manually or automatically determined area of the project. You can optimize and use larger tile sizes based on your available memory and the recommendation table in Performance Optimization.

ScenarioDefault tile size in GSD units
Default1000 x 1000
Aerial Oblique2000 x 2000
Aerial Nadir2000 x 2000

The following parameters are deprecated and will be removed in the next release. Furthermore, it is not possible to set a non-cubic 3D tile size.

CLI
-tilesize2d <value>
-tilesize3d <value> <value>


Initial terrain height

Specify mean initial terrain height for a priori image selection. This can also be disabled.

-initterrain <value>
-no_initterrain

Utilize initial terrain height information in order to select images involved in specified Key Parameters#Area. This way, images outside the specified area don't have to be considered and undistortion as well as image pyramid creation can be skipped for these images, which may significantly speed up the project preparation stage. When using Inpho project files, given terrain height information will be considered automatically.

Image Selection

Whenever the Project ROI is specified, such that it can be projected in the image space - that is, either a bounded volume (2D limitation + Z range), or a 2D region coupled with a specified terrain height - a prior selection of the input images that capture it is possible.

In the GUI, the user will be queried during Project Definition wizard, whether this selection should be performed. Enabling it can potentially accelerate the processing significantly, so this is recommended.

For the CLI, the flags bellow enable/disable this option. Note that the -initterrain flag enables it by default. If these flags are not passed at all and the Project ROI can be mapped in the image space, the image selection is enabled by default.

CLI
-apply_area_based_image_selection
-no_apply_area_based_image_selection


Point cloud format / LAZ compression

Activate LAZ compression for cloud outputs.

CLI
-laz

Enabling LAZ compression will significantly reduce the storage consumption of the cloud outputs (see table below). This will also affect the way in which point clouds are stored. When writing uncompressed LAS point clouds, each cloud tile will be populated with the triangulated points from all involved (multi-)stereo models. When LAZ compression is activated, however, there will be separate congruent cloud tiles containing only triangulated points from one specific stereo model each. Please also note that activating LAZ compression may significantly slow down cloud reading speed for subsequent processing steps, especially the DSM and Cloud Filter.

Dataset

Dense clouds size (las)2.9 GB1.5 GB10.4 GB
Dense clouds size (laz)743 MB341 MB3 GB
Compression rate74%78%71%




Filtered clouds size (las)402 MB345 MB2.5 GB
Filtered clouds size (laz)158 MB97.3 MB1.2 GB
Compression rate61%72%52%