Multiflight LiDAR data processing in TOPOLIDAR software

Introduction

Progress in the application of drones for high-precision survey is ongoing. Until a few years ago, it seemed that airborne laser scanning technology was the preserve of large companies with multimillion-dollar investments.

Since wide range of affordable LIDAR equipment from TOPODRONE has been realised,  an increasing number of surveyors, geologists, constructors, engineers are interested in UAV LiDAR technologies.

Many people ask the questions: How difficult or easy is it to process LiDAR data? What software is needed? Is this technology available and understandable for any surveyor?

Project Execution

To answer these questions, we have prepared an article about an actually completed project on engineering surveys for the reconstruction of an object located on an area of 120 ha, covered by dense forest and rugged terrain crossed by deep gullies and ravines.

Fig. 1. TOPODRONE LIDAR 100 LITE

TOPODRONE LIDAR 100 LITE mounted on the drone DJI MATRICE 200 was chosen as the main instrument for performing measurements.

We used an affordable and easy-to-operate TOPODRONE DJI MAVIC 2 PRO PPK survey drone to create orthophotos and colorised points clouds in RGB.

One of the main advantages of this equipment set is its relative compactness and mobility, as well as the possibility of its transportation in the luggage of any airline, which is very important for survey company which works in different parts of the world.

The field work including flight planning, base station installation, UAV laser scanning and aerial survey was done in one working day by two specialists. It is not difficult to imagine how much time and manpower would have been required if standard total station surveying methods had been used.

For mission planning we used the professional UGCS software, which has all the necessary tools to prepare not only aerial survey routes, but also LiDAR scanning, with the terrain following mode. The are was covered by just three LIDAR missions.

Fig. 2. Mission planning in UGCS software

Fig. 3. Drone controlling during flights, the smartphone screen displays the image from FPV camera, the laptop displays flight parameters and the position of the drone

After completing the field work, we began the data processing, all entire process of trajectories calculating  and LiDAR points cloud generating was performed in the field on an ordinary laptop within 15-30 minutes of the flights.

For this we used the TOPOLIDAR software, which has two basic modules:

• POST PROCESSING for high-precision trajectory calculation based on datasets from GNSS and IMU (inertial measurement) systems
• POINTS CLOUD GENERATION  to generate a point cloud

The choice of TOPOLIDAR software was determined by the following advantages: affordable cost of the software, usability and ease of interface, high processing speed, support for different formats of GNSS  data, availability of tools for automated, batch processing of multiple flights, both for high-precision trajectory calculation and for point cloud generation, flexible settings for point cloud creation.

To perform post-processing and trajectory equalization, we loaded static measurement data from the base station in RINEX format and entered its high-precision coordinates into WGS84, and loaded data from the drone (UBX and IMR files).

Then we activated the batch processing function and added two additional data sets.

Fig. 4. TOPOLIDAR software interface.

Fig. 5. Loading data for automated processing of multiple data sets at once

As a result, the software calculated a high-precision trajectory of three drone flights with a coordinate measurement frequency of 200 Hz (200 times per second).

Fig. 6. Results of trajectory calculation

At the next step we started the process of point cloud generation by selecting the output projection, altitude system, as well as the necessary part of the route for processing (Fig. 8) and entered calibration information individual for LiDAR system. 

it should be noted that it is possible to add different types of projections and geoids by the user.

To reduce the time, we used the very handy batch processing function and loaded two additional routes for the calculation.

Fig. 7. Selecting the coordinate system and geoid to generate the point cloud.

Fig. 8. Selecting a part of the trajectory to generate a point cloud

The point cloud generation process did not take long, on average it takes 2-3 minutes  to calculate the data from one 30 minutes flight.

Thus, after 15 minutes of data processing we have generated the data of airborne laser scanning for an area of about 120 ha presented in Fig. 9-15, as well as performed automatic classification of the terrain and construction of contour lines.

Fig. 9. LiDAR points cloud

Fig. 10. LiDAR points cloud

Fig. 11. LiDAR points cloud showing power lines, poles, buildings and vegetation

In the next step, we performed colorising of LiDAR data in RGB using the images obtained with TOPODRONE DJI MAVIC 2 PRO PPK. The advantage of TOPODRONE technology is the ability to combine geo-referenced images from any drone equipped with PPK or RTK system with laser scanning materials to obtain a point cloud in RGB.

Fig. 12. Point cloud in RGB colors.

Fig. 13. Point cloud in RGB colors.

Fig. 14. Point cloud in RGB colors.

Fig. 15. Point cloud in RGB colors.

We were impressed by high density of reflections obtained under the trees, which made it possible to determine the terrain level with high accuracy, to perform automatic classification and to create contour lines. The digital terrain model and counter lines are shown in Fig. 15-17.

Fig. 18-20 show "sections" of airborne laser scanning data, by which the surface of the ground, crowns and trunks of trees, shrub vegetation can be easily deciphered.

As practice has shown, the density of LIDAR data under trees and high vegetation, obtained by TOPODRONE equipment substantially exceeds the data obtained from the widely advertised solid-state lidars of other manufacturers, about what we will write the detailed report in our future articles.

Fig. 16. Digital elevation model.

Fig. 17. Contour lines.

Fig. 18. Contour lines constructed in automatic mode, according to the classified point cloud.

Fig. 19. A section of the point cloud, clearly showing the density of reflections under trees and shrubs.

Fig. 20. A section of the point cloud that clearly reads the landforms.

Fig. 21. Points cloud profile and terrain level under dense shrub vegetation.

Conclusion

The results of this project clearly show the accessibility of airborne laser scanning technology to almost any surveyor with skills in drone control, post-processing GNSS measurements and working with GIS/CAD applications.

The use of the TOPODRONE LIDAR 100 LITE system opens up the hitherto inaccessible world of LiDAR high technology for a wide range of survey companies and small survey crews, arming them with an excellent and mobile tool in the tough, competitive market for surveying services today, significantly reducing financial costs and time of work.

Below is online access to the resulting data, where you can take a fascinating virtual journey and appreciate the quality of the data, especially under the crowns of century-old trees.

Map

3D