«We don’t use any ground control points for LiDAR data processing, georeferencing of points cloud, strip alignment etc. We need just few check points to evaluate the accuracy».
To prove this statement, we would like to share the results of drone LiDAR mapping project carried out in Riga, Latvia together with METRUM company over the area with deep vegetation, power lines, roads and buildings.
The main idea of this project was to show in real conditions TOPODRONE LiDAR equipment capabilities and to check the accuracy of drone LIDAR point cloud in difficult for using common mapping technologies areas like canopies, deep vegetation, roads, power line towers.
To evaluate the quality of 3D LiDAR scanning more than 200 check points were measured by survey grade RTK receiver and total station.
Fig. 1. Check points locations.
Before the flight we installed GNSS receiver as a base station. It is enough to setup static observation with 1 Hz rate. What you need is just to measure precise position of the base.
Fig. 2. Base station.
Laser scanning mission was prepared in UgCS Expert software directly in the field.
Fig. 3. LiDAR mission planning in UgCS software.
The area of 16 hectares of dense forest was surveyed by TOPODRONE LiDAR 100 LITE installed on DJI Matrice 300 within 12 minutes.
As you can see in photos we made survey in the evening and it is the greatest advantage of LiDAR technology that it doesn’t depends on light conditions and you can capture data even at night.
After the flight precise trajectory was post processed in TOPODRONE Post Processing software and point cloud was generated within 2-3 minutes in Latvian projection LKS-92 TM and LV14 GEOID with using of precalibrated LiDAR parameters. It took us no more than 15 minutes to accomplish data processing.
Fig. 4. GNSS & IMU data post processing in TOPODRONE software.
Fig. 5. Selecting necessary part of trajectory for future point cloud reconstruction.
Fig. 6. Real time point cloud generation.
Fig. 7. LiDAR point cloud captured by TOPODRONE LIDAR 100 LITE.
Next day photogrammetry mission was performed in good light conditions to capture photos to colorise LiDAR points cloud by DJI Phantom 4 Pro drone with additionally installed TOPODRONE DJI Phantom 4 Pro PPK Upgrade Kit.
Fig. 8. TOPODRONE DJI Phantom 4 Pro PPK was used to colorise LiDAR point cloud.
Fig. 9. RGB point cloud.
Fig. 10. 3D model of the tower.
This project shows real benefits of using LiDAR technology in comparison with photogrammetry in order to get accurate and reliable results for vegetation areas in a short time.
Fig. 11. Check points location.
Fig. 12. Check point location in the forest.
Fig. 13. Check points location.
Automatic accuracy report
|
ID |
X |
Y |
Known Z |
Z |
dz |
|
1 |
513090.953000 |
309903.726000 |
11.180000 |
11.215600 |
0.035600 |
|
4 |
513095.226000 |
309906.461000 |
11.270000 |
11.241000 |
-0.029000 |
|
6 |
513098.873000 |
309908.183000 |
11.230000 |
11.209700 |
-0.020300 |
|
7 |
513100.787000 |
309908.760000 |
11.070000 |
11.108000 |
0.038000 |
|
10 |
513102.815000 |
309908.853000 |
11.230000 |
11.220000 |
-0.010000 |
|
12 |
513104.831000 |
309908.686000 |
11.230000 |
11.250300 |
0.020300 |
|
13 |
513106.818000 |
309908.270000 |
11.080000 |
11.138300 |
0.058300 |
|
14 |
513106.808000 |
309908.232000 |
11.240000 |
11.220000 |
-0.020000 |
|
15 |
513109.638000 |
309907.212000 |
11.090000 |
11.128000 |
0.038000 |
|
16 |
513109.637000 |
309907.157000 |
11.240000 |
11.199000 |
-0.041000 |
|
18 |
513112.363000 |
309905.830000 |
11.220000 |
11.168700 |
-0.051300 |
|
19 |
513115.012000 |
309904.296000 |
11.060000 |
11.118000 |
0.058000 |
|
21 |
513116.716000 |
309903.034000 |
11.030000 |
11.036300 |
0.006300 |
|
23 |
513118.598000 |
309900.742000 |
10.980000 |
10.975300 |
-0.004700 |
|
25 |
513119.142000 |
309899.281000 |
10.960000 |
10.996000 |
0.036000 |
|
27 |
513119.540000 |
309895.429000 |
11.000000 |
10.965300 |
-0.034700 |
|
31 |
513124.539000 |
309895.856000 |
10.970000 |
10.955000 |
-0.015000 |
|
32 |
513125.316000 |
309899.336000 |
10.920000 |
10.894000 |
-0.026000 |
|
33 |
513127.154000 |
309904.267000 |
10.920000 |
10.914000 |
-0.006000 |
|
34 |
513130.334000 |
309907.784000 |
11.020000 |
11.036000 |
0.016000 |
|
35 |
513130.852000 |
309908.335000 |
11.030000 |
11.035700 |
0.005700 |
|
36 |
513130.952000 |
309908.160000 |
11.210000 |
11.208700 |
-0.001300 |
|
37 |
513141.572000 |
309905.624000 |
11.140000 |
11.198300 |
0.058300 |
|
39 |
513144.383000 |
309904.634000 |
11.190000 |
11.228700 |
0.038700 |
|
43 |
513144.612000 |
309903.617000 |
11.400000 |
11.422000 |
0.022000 |
|
59 |
513122.341000 |
309907.410000 |
10.830000 |
10.772000 |
-0.058000 |
|
60 |
513123.038000 |
309925.571000 |
11.290000 |
11.341000 |
0.051000 |
|
62 |
513123.152000 |
309925.547000 |
11.420000 |
11.392000 |
-0.028000 |
|
63 |
513123.268000 |
309924.831000 |
11.280000 |
11.310300 |
0.030300 |
|
65 |
513123.773000 |
309924.179000 |
11.280000 |
11.300300 |
0.020300 |
|
67 |
513124.398000 |
309923.698000 |
11.270000 |
11.280000 |
0.010000 |
|
69 |
513125.135000 |
309923.444000 |
11.280000 |
11.300300 |
0.020300 |
|
71 |
513125.915000 |
309923.392000 |
11.280000 |
11.310700 |
0.030700 |
|
73 |
513127.985000 |
309923.977000 |
11.270000 |
11.280000 |
0.010000 |
|
75 |
513127.941000 |
309924.127000 |
11.430000 |
11.381700 |
-0.048300 |
|
76 |
513130.820000 |
309925.949000 |
11.300000 |
11.289700 |
-0.010300 |
|
78 |
513130.790000 |
309926.092000 |
11.440000 |
11.421700 |
-0.018300 |
|
79 |
513124.613000 |
309925.316000 |
11.470000 |
11.503700 |
0.033700 |
|
82 |
513130.603000 |
309931.773000 |
11.500000 |
11.543700 |
0.043700 |
|
84 |
513142.618000 |
309936.374000 |
11.470000 |
11.421300 |
-0.048700 |
|
85 |
513142.637000 |
309936.345000 |
11.360000 |
11.390700 |
0.030700 |
|
87 |
513148.158000 |
309929.958000 |
11.370000 |
11.340000 |
-0.030000 |
|
88 |
513148.227000 |
309929.906000 |
11.370000 |
11.340000 |
-0.030000 |
|
89 |
513149.285000 |
309928.719000 |
11.420000 |
11.391000 |
-0.029000 |
|
94 |
513136.554000 |
309919.675000 |
11.310000 |
11.289700 |
-0.020300 |
|
95 |
513136.586000 |
309919.596000 |
11.300000 |
11.300000 |
0.000000 |
|
96 |
513137.714000 |
309918.374000 |
11.330000 |
11.310000 |
-0.020000 |
|
97 |
513136.267000 |
309916.336000 |
11.220000 |
11.218300 |
-0.001700 |
|
99 |
513133.512000 |
309916.969000 |
11.290000 |
11.259000 |
-0.031000 |
|
100 |
513133.583000 |
309916.855000 |
11.290000 |
11.269000 |
-0.021000 |
|
101 |
513132.769000 |
309916.398000 |
11.140000 |
11.116700 |
-0.023300 |
|
102 |
513132.726000 |
309915.577000 |
11.110000 |
11.137300 |
0.027300 |
|
104 |
513132.758000 |
309914.814000 |
11.110000 |
11.106700 |
-0.003300 |
|
106 |
513133.393000 |
309913.503000 |
11.080000 |
11.076300 |
-0.003700 |
|
108 |
513134.387000 |
309912.621000 |
11.050000 |
11.056000 |
0.006000 |
|
109 |
513134.392000 |
309912.652000 |
11.090000 |
11.056000 |
-0.034000 |
|
110 |
513134.452000 |
309912.708000 |
11.090000 |
11.056000 |
-0.034000 |
|
116 |
513136.506000 |
309911.546000 |
11.090000 |
11.045700 |
-0.044300 |
|
170 |
513257.893000 |
310026.660000 |
11.660000 |
11.601000 |
-0.059000 |
|
176 |
513249.628000 |
310020.407000 |
11.750000 |
11.733700 |
-0.016300 |
|
179 |
513249.083000 |
310021.116000 |
11.760000 |
11.754000 |
-0.006000 |
|
180 |
513248.514000 |
310021.394000 |
11.750000 |
11.770200 |
0.020200 |
|
182 |
513247.664000 |
310021.191000 |
11.750000 |
11.788400 |
0.038400 |
|
186 |
513247.307000 |
310015.484000 |
11.510000 |
11.489700 |
-0.020300 |
|
198 |
513247.161000 |
310029.355000 |
11.860000 |
11.855300 |
-0.004700 |
|
201 |
513247.283000 |
310030.800000 |
11.860000 |
11.865300 |
0.005300 |
After that point cloud was colorized and we started to evaluated accuracy by several check points which were measured under trees, on edges of road curbs and on power line tower.
As you can see from automatic control point report we achieve the excellent accuracy within 3-5 cm and NO GROUND POINTS were used for strip alignment and point cloud adjustment.
Fig. 14. Drone and terrestrial LiDAR scanning data.
To finalize TOPODRONE LiDAR 100 LITE accuracy evaluation, we performed drone and terrestrial LiDAR scanning data comparison in a dense forest.
Fig. 15. Drone and terrestrial LiDAR scanning data comparison.
The results show excellent convergence of the data. Terrestrial scanning provides more details in the lower part of the tree crowns, but does not allow to determine the actual height.
Drone laser scanning allows us to reliably calculate both the number of trees and the height and diameter of trunks.
The outstanding results of this project show real advantages of LiDAR technology in comparison with photogrammetry such as:
- Efficiency. You will need less overlap to generate 3D LiDAR maps.
- Fast results delivery. Drone LiDAR mapping is much easier and faster in comparison with photogrammetry survey.
- High mapping accuracy. TOPODRONE LiDAR system provides with stable 3-5 cm XYZ LiDAR mapping accuracy due to the well calibrated Velodyne sensors and high precision Honeywell IMU.
- Survey at any time. LiDAR scanning doesn’t depend on light conditions, and it is possible to perform survey even at night.
