Import Heron data

To import the Heron® data as it is after being exported from Heron® Desktop.


The transition from Heron® Desktop to Reconstructor® is guarantee by two main steps:

STEP 1 | in HERON® Desktop 

saves the acquired points according to user defined parameters and colorizes (only for Color systems) them.

STEP 2 | in Reconstructor® 

applies dedicated filters to the point clouds, generates the Blueprints and  organizes points and images (streaming and single-shots) to be navigable along the trajectory using the Blueprint Navigator.


Different parameters configurations can be used in HERON® Desktop and Reconstructor® to run the import process according to the desired time/result ratio:

  • Very fast
  • Fast
  • Balanced
  • High resolution



STEP 1 | in HERON® Desktop 


The correct way to transfer the Heron® data from HERON® Desktop to Reconstructor® is to use the Go To Reconstructor command in the HERON® Desktop toolbar.

The GoTo Reconstructor dialog is subdivided in two sections:


A - SOURCES


The settings in this sections allow:

  1. To export points from both the sensors (for Heron MS Twin) or only from one sensor.
    For HERON MS Twin data : is the horizontal sensor, is the oblique one.
    For Heron LITE (or AC) data: is the only available sensor.

  2. To apply a min/max range filter (distance of the measured points from the HERON capture head).

  3. To filter out the images in correspondence of large rotation along the trajectory; the filtered images are not used during the point cloud colorization (360° for using all the captured images; < 360° to exclude images during the colorization, where the HERON capture head has large rotations). This possibility is valid only for Heron MS or AC  systems (the automatic colorization of Heron LITE system is not available).

  4. To remove points with low confidence. The confidence is calculated during the Odometer step: 
    0 = all points are exported
    99 = only points with 100% registration quality are exported



B - OPTIONS


The settings in this sections allow:


  1. To voxelize the LiDAR sensor data. By selecting the cell size for the voxelization (Voxel size value),you can decide the resolution of the point cloud. Example: setting the voxel size to 0.03 m, a point every 3 cm will be selected to generate the final point cloud.

  2. The Average points option determines the way of selecting the points in a voxel. If checked [suggested], each point is computed as the average of all the points in a voxel: if unchecked, the first point along the trajectory fitting in the voxel is selected.

  3. Setting the Memory cap option you can set the amount of memory used in the process according to the PC's resources.

  4. The Clean Data checkbox allows you to apply the filter file obtained from the the Clean Data step. 



STEP 2 | in Reconstructor®


Once the Go to Reconstructor process in finished, close HERON® Desktop Mapper and go back to the launch window, then press Reconstructor in the top toolbar.



A Reconstructor® project (linked with the current HERON® Desktop project) is automatically created.

In the HERON top toolbar menu of Reconstructor® press the Import Heron button to open a wizard window that guides you along the import phase.



Import Global Optimization/Single Trajectory

Click on this button to import the results of a Global Optimization as they are exported from HERON® Desktop and saved in the Heron workspace's folder exports/reconstructor.

Select one Global Optimization from the proposed list and click on OK.




The wizard window is divided in 6 main sections:


  1. Point Clouds tree items
    It gives the access to single items saved after the global optimization step in HERON® Desktop (point clouds, trajectory, images and tags).

  2. Cloud data
    After selecting a point cloud item, the color layers you want to keep, parameters and setting for importing, voxeling and filtering the 3D point clouds are selectable.
    A Voxelization of the imported data can be applied, where the best aligned point is selected inside each voxel.
    A Clean geometry filter can be applied to reduce the LiDAR noise without reducing the point cloud density. 
    Outliers filter is a noise removal filter that permits to delete the lonely points outside a detected surface.
    The normals per each 3D point can be enhanced providing coherent inclination for each surface (important and useful when using fitting tools inside Reconstructor® or in third parties software - e.g. Edgewise®, Verity® to automatically detect geometrical features as walls, floors, ceilings -).
    The Statistical outliers remover (SOR) filter removes outliers in order to obtain cleaner data.
    The Hide black points filter hides all the points in the cloud that are colored in full black, i.e. RGB components [0 0 0].
    3D Reflectance is a special color layer that mixes LiDAR intensity and point normals (inclination) to enhance both geometrical and surface characteristics.

    See more details about the point clouds processing at Cloud Processing  and in particular at Pre-processing Point Clouds entries.

  3. Camera
    It includes the setting for importing images captured by Heron®.

  4. Control Points
    It defines the folder including the control points coordinates.

  5. Tags
    It includes the setting for importing tags.

  6. Blueprint
    It automatically generates a blueprint.



The tree structure including all the items saved in the Global Optimization is so displayed.

By clicking on a single item it is possible to edit its import settings (for details see Workflows - Parameters configurations).


By default the five sections from 2 to 6 are providing with a "medium" preset:

2. Assuming that in HERON® Desktop the survey is saved without any voxelization, Reconstructor® applies:

    • a voxel size of 2 cm. The best aligned point is selected inside each voxel.
    • a Black Points hiding
    • the computation of 3D Reflectance, a special color layer that mixes LiDAR intensity and point normals (inclination) to enhance both geometrical and surface characteristics.

These filters are applied only to Heron® data, not to any other cloud present in the Global Optimization.


3. The camera calibration is automatically applied on single-shot images to guarantee an optimal overlap to the 3D point clouds.

It is really recommended to keep flagged the Camera calibration process.


4. Flag "Control Points" and choose, between proposed options, the folder including the Control Points saved in the linked HERON® Desktop project.
Every imported Global Optimization will have its own User Coordinate System.

If working with pole tags, each list points to its own UCS and if working with manual tags only one default UCS is necessary.


5. Tags acquired on the field are imported by default.


6. A blueprint of the surveyed clouds is automatically created and linked to trajectory, points and images.




Advanced parameters


To access the advanced parameters, select the specific items from tree structure [Section 1]. 

Global Optimization

Includes all the objects in the Global Optimization you want to import.

Single Tracklet

It is related to the a single tracklet, composed by a point cloud, a trajectory, cameras (if present) and tags.

Point Clouds

The point clouds in the Global Optimization can be both unstructured (e.g. Heron® data) and gridded (e.g. static scans).

Trajectory connected to the relative point cloud

If the point cloud is the Heron® mobile data, a trajectory is directly connected to it.

Cameras

If present, the camera images are connected to the trajectory (both single-shot and streaming images, split in 2 different headings).

In latest Heron systems, the streaming cameras are used to colorize the imported point clouds (a RGB color layer is created on the basis of streaming pictures).

The folder containing the cameras and their calibration in the HERON® Desktop project is automatically linked, as well as the camera properties are automatically recognized.

This is valid both for single-shot (Highest Resolution Images) and streaming images data set. The position of the pictures is automatically recognized.

The cameras will be visible in the Blueprint Navigator.


By selecting the camera items, the advance parameters for images import are reported. First of all the camera model and the directory containing the pictures. Then the calibration file position and the camera time offset to correctly set the calibration of cameras on LiDAr data.
The “first camera” in the list is referred to streaming images, while the “second camera” in the list is referred to single-shot/high resolution images.

Flag the camera calibration to make the calibration effective during the data import phase.

Don't change these parameters without contacting Gexcel!

Tags

Both annotations and targets are here included.


By selecting the Tags item, the following advance option appears


By pressing the Configure Tags button a list of all the tags is showed and here you can choose to maintain or change the role of each tag inside Reconstructor®.


Check to don't import the tag in the Heron® Global Optimization/Survey.

Check to consider the tag as an annotation.

Check to consider the tag as a target (to be used for georeferentiation and registration). 



Blueprint

By checking Automatically generate blueprint a blueprint related to the point cloud(s) you are importing is automatically created. Its resolution is depending on the dimensions of the point clouds bounding boxes.

 This automatic blueprint is generated on a plane parallel to the XY plane of the current UCS (at the time of the import process). 


To manage the blueprint layers see Create/Edit Heron Survey.



At the end of the process, a Heron® Survey including all the imported items will be inserted in the current project.


Note

If no Heron® Desktop project is already connected to the Reconstructor® project, by clicking on the Import Heron button the ws_info.json file included in the desired Heron® Desktop project's folder is requested. After this point there is the same process.