Lidar Processing Settings
The processing tab contains lidar processing settings associated with the project.
FOV and Range are the most commonly adjusted parameters. With aerial data, a 90 degree FOV is typically a good compromise between swath size and quality returns (returns far from nadir are typically noisy). A minimum range can help remove points from aircraft landing gear.
Returns to fuse controls which returns are fused into the pointcloud. With multi-return lidar sensors, such as Riegl lidars, multiple returns are recorded for each pulse emitted, and this return index is stored into the data (RXP or SDCX file).
Downward FoV specifies the swath width in degrees centered on vehicle-nadir. (90 degrees is 45 degrees on either side of nadir).
Fuse Returns from every nth shot and nth line. This can be used to fuse only a subset of points, which is useful for expediting processing time when acquired density is very high. See an example below section on downsampling.
Valid Intensity is a scaled intensity value - very low reflectance values, less than -25 dB, may likely be noise.
Valid Deviation refers to how much a return waveform shape deviates from an ideal, leptokurtic return shape.
Minimum and maximum range - Adjust to eliminate points that are either very close or very far from the sensor.
Returns to fuse, Valid Intensity, and Valid Deviation are not applicable to all LiDAR scanners.
Downsampling with Line Scanners
If you do not require the full recorded lidar density, you can limit which points are fused by adjusting the Fuse returns from every nth shot and nth line.
Examples
Fusing the 1st shot from every 10th line will result in fusing only 10% of the density, as only 1 out of every 10 scan lines are fused.
Fusing the 2nd shot of every 1st line, will result in fusing only 50% of the density, as every scan line is fused, however every other shot is skipped.
In the example below, only about 3% of the data will be fused, which will drastically speed up processing times, in the case of a field data check or other cursory examination of the pointcloud:
Blocking Out Frustums
In some payload configurations, the LiDAR sensor can be mounted in such a way that it might constantly scan parts of the vehicle or craft, (i.e. a portion of a car’s roof or a UAV’s landing gear). This will create reflections in the resulting 3D point cloud. There are two options to prevent this from happening:
Set the minimum range value to a value greater than difference of the distance from the LiDAR sensor to the obstacle.
Block certain frustums, which are angular regions. Up to four regions can be blocked at the same time.
In order to calculate frustums, you must know the orientation of the LiDAR sensor as it is mounted. For a Velodyne sensor, it is indicated by the cable being 180 degrees. For a RIEGL sensor, it is indicated by R logo on the front of the sensor.
You then select a horizontal min/max and a vertical min/max of the range you would like to block as well as a min/max distance range. RIEGL sensors do not have a vertical min/max option because they are line scanners and therefore do not have a vertical field of view. Horizontal refers to the 360 degree rotation of the scanner, so essentially to the left/right of your vehicle. Vertical refers to forward and backward.
To block out frustums you need to set each of the 3 parameters (Horizontal, Vertical, and Range) to include the points you wish to block. In this example below, we are blocking everything that is to the Right and Behind the vehicle. Points are masked that meet all criteria for the mask: Horizontal 330 to 360, and Vertical 0 and +45, and range 0 to 500.
You might need to set up more than 1 filter such as using one frustum to block from 0 to 90 degrees and another to block from 180 to 270 degrees. For example, if you want to preserve only 40 degrees, +20 and -20 from nadir, then you will need to set up 2 filters. First filter should block points if they are 0-160 degrees horizontal, and 0-1000 (or whatever your maximum range is for the scan) meters range. The second should block points if they are 200-360 degrees horizontal and 0-1000 meters range.
It is recommend adjust these parameters and inspect them in SpatialExplorer before fusing them and exporting your point cloud to a LAS/LAZ.
Last updated
