Automatic Estimation of Excavation Volume from Laser Mobile Mapping Data for Mountain Road Widening II

Artículo patrocinado por Extraco, Misturas, Lógica, Enmacosa e Ingeniería InSitu, dentro del proyecto SITEGI, cofinanciado por el CDTI. (2012). 

Article sponsored by Extraco, Misturas, Lógica, Enmacosa and Ingeniería Insitu inside the SITEGI project, cofinanced by the CDTI. (2012)

Continue from: http://carreteras-laser-escaner.blogspot.com/2014/11/automatic-estimation-of-excavation.html

3. Implementation and Testing of the Method
3.1. Software Implementation

The methodology described above is implemented on an ordinary Dell desktop computer, which has an Intel Xeon 3.6 GHz CPU on board and 16 GB random memory. The implementation of the software is in the C++ language. Also, the Point Cloud Library (PCL) statistical outliers filtering tool is used in the processing. The whole processing took 23.184 s for the tested dataset. 

3.2. Data Description

The point cloud data studied in this paper was acquired by the University of Vigo, Spain. The approximate location of the studied road is shown in Figure 7. 

The entire study area is shown in Figure 8. The study area contains a road in a mountainous region. An overhead view of the dataset is shown in Figure 8a. The mobile LiDAR system selected for this work was the Lynx Mobile Mapper from OPTECH. The Lynx uses two LiDAR sensors to collect survey-grade LiDAR data at 500,000 measurements per second with a 360° FOV (per scanner) [33,34]. 

The system incorporates the POS LV 520 unit produced by Applanix, which integrates an Inertial Navigation System with two GNSS antennas, providing an accuracy of 0.015° in heading, 0.005° in roll and pitch, 0.02 m in the X, Y dimension and 0.05 m in the Z axis. All those data are determined by differential GPS post-processing after data collection using GPS base station data. The coordinate system used for this work is UTM-WGS84. The original point cloud dataset contains 5,838,794 points and has an average point density of 2084 points per square meter. It covers a 132-m stretch of road. In Figure 8c, we can see that the road was constructed in a mountainous area and has steep embankments on either side.

Figure 7. Approximate location of the studied road.

Figure 8. Original point cloud dataset of the study area. (a) Original laser mobile mapping systems (LMMS) point cloud data in 3D view; (b) Original LMMS point cloud data in 3D side view; (c) Study area in Google Street View; (d) Data collection with LMMS.

3.3. Geometric Computations
3.3.1. Slope Computation

As depicted in Figure 9, the slope in this area varies from 0 to 88.1 degrees. The figure is colour coded with red indicating a large slope and blue indicating a small slope. Points with small slopes are mainly road points. The points of the road have larger slope values corresponding to the steepness of the road side. The dots on the road encircled in red have larger slope value than other road points; these are in fact traffic cones that can also be seen in Figure 8d. There is a known landslide site located within the light green circle, which has a smaller slope value that stands in contrast to the steep roadside of its neighboring points.

Figure 9. 2D slope at each point of the studied road.

3.3.2. Road Detection and Segmentation

Road points were identified according to the method outlined in Section 2. The minimum virtual grid size was set to 0.1 m, because there is a very high point density in the original point cloud dataset. Additionally, the height threshold was set to 0.3 m and the angle threshold was set to 15 degrees. In this processing, a total of 42,717 road points was abstracted and segmented, as shown in Figure 10.

Figure 10. Segmented road from the original point cloud dataset.

3.3.3. Volume Computation

First, based on the segmented road points, the road outline was determined following the method of Section 2.5. In this paper, the local descriptor threshold was set at 1.5, which means that all points with an edge descriptor value greater than 1.5 were regarded as road outline points. 

The abstraction results are shown in Figure 11. In total, 429 outline points were identified. 

Figure 11a depicts the abstracted road central line, and Figure 11b illustrates in addition the road lines of a projected road expansion by 4 m on each side.

After the extraction of the points above the possible location of the widened road (i.e., 4 m left and right of the current road), the vertical distances between points representing the current surface and the expanded road planes were determined, and the excavation volume was estimated. Figure 12 shows two profiles of the current surface height at a distance of 4 m left and right from the current roadsides. 

The horizontal axis follows the road starting from its lowest point. In this paper, the resolution in the road parallel direction was set to 1 m and in the road perpendicular direction to 0.5 m. Figure 12 shows the overall height increase of the surface profiles in the road parallel direction.

Figure 13 shows the volumes of the slices which were computed as described in Figure 6. Because of some low water drain elements on the road side, there are values that are below 0, which means that if the road is widened by 4 m, some of the volume should be moved to those locations to match the road surface height. The arrow points to a location that has a greater surface height resulting in a slice with a larger volume.

Figure 11. Road outline, central line and expanded road outline. (a) Road central line and road outline; (b) Additional expanded road outline.

Figure 12. Profiled height on the expanded roadside.

Figure 14 shows the cumulative volume in the road parallel direction. At the left side of the road, 542.22 m3 needs to be excavated compared to 462.35 m3 on the right side of the road. Because the studied road is not straight, the road parallel direction distances for the two road sides differ. The right side is 137.2 m long, whereas the left side is 124.1 m long. At the location indicated by the arrow in both Figures 13 and 14, there is one slice with a particularly large volume which causes a steep rise in the cumulative volume. As shown in the cumulative volume estimation results, the material volume that needs to be excavated on the left side is 8% greater than on the right side. When applied in real applications, such results may help engineers to optimize road widening design to minimize the time and costs of the project.

Figure 13. Slice volume computed from the expanded road outlines on both sides.

Figure 14. Cumulative volume on the extended roadside.

4. Quality Discussion and Validation

Contiue on: http://carreteras-laser-escaner.blogspot.com/2014/11/automatic-estimation-of-excavation_19.html