For measurement, LIDAR uses laser beams that transform the real world into a virtual 3D environment by capturing millions of points per second. The following types of scanners are now distinguished: ALS (Airborne Laser Scanning), TLS (Terrestrial Laser Scanning), MLS (Mobile Airborne Laser Scanning), and HLS (Handheld Laser Scanner). A key element for each measurement is to select an appropriate scanner, since one type of scanner cannot replace another one. Nevertheless, regardless of the selection, the principle of operation is always the same: a point with coordinates XYZ is recorded for each point in space which is hit by the laser beam. The scanner measures the angle and distance between the instrument and the object being measured. At the same time, the scanner records the intensity rate and the reflection of the laser from the surface being measured. Currently, most scanners have additionally built-in cameras, which provides an RGB color value for each point and in turn this ultimately allows the point cloud to be colored.
In a typical project, to provide a complete data set, the object being measured must be captured from different places. With large objects, such as buildings, we usually scan at a density of 3mm per 10m. In practice, this means a 3 millimeter spacing between each point measured at a distance of 10 m from the scanner. If the distance from the scanner’s position to the object is increased, this density will decrease, i.e. at a distance of 20m we will obtain a point density of 6mm. This is a key parameter that is determined before starting each measurement.
To combine all scans, a network of relationships between them must be created. This is usually done by capturing the same points on the object, using special signals in the field of view or by fitting a cloud into a cloud, if the scans overlap sufficiently.
After the data are downloaded into the computer, they can be processed in different ways based on the point cloud. The essential issue is to create spatial solids of the objects.
At the beginning, erroneous points or disturbances should be removed from the model as well as the overall accuracy and compliance with the assumed tolerance should be checked. Using appropriate transfers, we convert the point cloud to the virtual environment of the target application where we can create the desired model of the object using appropriate tools.
At this stage, we check again the quality of the data acquired, having first cleaned them of any unnecessary points, and the output data format ordered by the client is prepared. If you have any interesting project in your mind, please contact us.
The output of scanning is a homogenous point cloud for the entire object, from which we create at a later stage:
Add us to favorites and keep to date with the last news