Kinematic Laser Scanning

Quality Analysis of Point clouds

The uncertainty of parameters derived from point cloud data is hard to derive since the uncertainties of the single components of the point cloud generation are unknown or cannot be modeled properly. Especially, the correlations between every single point are unknown but very important in order to achieve plausible uncertainties of the extracted parameters in a forward modeling approach. Alternatively, empirical approaches can help us to characterize the uncertainty of these parameters. The idea is to extract relevant parameters from common applications directly from the point cloud data.By comparing them to reference values we can determine the accuracy of these parameters.

There are three major questions arising:

• Can we find standardized parameters for the evaluation?
• Are the derived accuracies transferable to other locations?
• How do we obtain the reference information?

The goal of this research is to quantify the uncertainty of the point cloud for common applications, in order to make a statement, whether the accuracy of the system is sufficient to solve the task we have to tackle.

Contacts:

• M.Sc. Markus Wagner
• Dr.-Ing. Berit Jost
• PD Dr. rer. nat. Lasse Klingbeil
• Prof. Dr. Heiner Kuhlmann

The Potential of UAV-based Laser Scanning for Deformation Monitoring

The use of unmanned aerial vehicles (UAV) in monitoring applications is constantly increasing due to the improvement in sensor technology and the associated higher accuracy that can be achieved. As a result, UAV-based laser scanning is already being used in various deformation monitoring applications such as the monitoring of landslides or land deformations. The main challenges, which also limit the accuracy of the resulting georeferenced point cloud are given by the trajectory estimation, the measurement environment and the flight planning. Difficult conditions and high accuracy demands are especially given for the monitoring of a water dam.

While the use of area-based measurements such as terrestrial laser scanning (TLS) is an already established approach for such monitoring tasks, the use of a similar technology on a platform such as a UAV is promising and investigated in this work by acquiring measurements at a water dam. In addition to the proposal of a flight pattern for the measurements, the trajectory estimation results are evaluated in detail. Due to critical GNSS conditions, positioning errors lead to systematic shifts between single flight strips. Subsequent optimization with known control points allows the point cloud to be compared to a TLS reference.

Contacts:

• M.Sc. Ansgar Dreier
• PD Dr. rer. nat. Lasse Klingbeil
• Prof. Dr. Heiner Kuhlmann

Quality Assessment of UAV-based Laser Scanning

The rapid development of sensor technology in the field of airborne laser scanning meanwhile offers the possibility to operate compact profile line scanners on an Unmanned Aerial Vehicle (UAV). The advantages of using a combination of UAV and laser scanner are many, and benefit in particular from the coverage of large areas in a relatively short time.

After pose estimation, the combination of the trajectory with the corresponding measurements from the laser scanner results in a georeferenced 3D point cloud. The synchronized acquisition of camera images additionally provides the possibility to colorize the estimated point clouds.

In the context of this UAV-based laser scanning system, our research deals with various areas such as the quality analysis or the use of laser scanning for high-throughput phenotyping. Related to the quality analysis, our investigations include aspects such as the absolute accuracy of the processed point cloud depending on the measurement configuration and flight parameters. 

Contacts:

• M.Sc. Ansgar Dreier
• PD Dr. rer. nat. Lasse Klingbeil
• Prof. Dr. Heiner Kuhlmann

High Resolution Mobile Laser Scanning

Mobile laser scanning has become an established measuring technique that is used for many applications in the fields of mapping, inventory, and monitoring. We developed a custom mobile mapping system based on a high resolution profile scanner, and navigation grade inertial measurement system.

The main challenges in the mobile laser scanning are the calibration of the multi-sensor system and the accurate estimation of the vehicle trajectory. Both are needed to transform the sensor readings from the scanner into a global reference frame.

We are developing methods for calibration of the system and we also developing algorithms for accurate trajectory estimation. With this we are able to create very dense and accurate point clouds.

Contacts:

• M.Sc. Markus Wagner
• PD Dr. rer. nat. Lasse Klingbeil
• Prof. Dr. Heiner Kuhlmann

Quality investigation of Mulit-target Capability

The multi-target detection capability of a laser scanner is based on full-waveform signal processing. This capability is used in the time of flight measurement technique where the full waveform of the reflected signal is recorded by the scanner. By the use of algorithms for the detection of peaks within the waveform, the corresponding range measurement is derived.

Therefore, the measurement of a single laser beam can detect multiple reflections from different objects. The capability of detecting multiple targets is commonly used in the application of forestry, where the information of different returning pulses can be used to capture the canopy and ground at the same time. The limitation imposed by the sensor design is usually referred to as the ‘vertical dead zone’, in which the systems are unable to separate two subsequent returns.

Contacts:

• M.Sc. Ansgar Dreier
• PD Dr. rer. nat. Lasse Klingbeil
• Prof. Dr. Heiner Kuhlmann