Airborne LiDAR has been making headlines in the last few months, with the Environment Agency announcing exciting plans to map the whole of England by 2020. To map on such a large scale, the equivalent of 32 million football pitches, the organisation will be using LiDAR (Light Detection and Ranging) sensors mounted on aircraft, a practice also known as Airborne Laser Scanning.
Airborne Laser Scanning (ALS) is conducted by using a LiDAR sensor mounted on an aircraft which can be either manned or unmanned. The sensor sends invisible near-infrared light pulses to the ground and the distance is calculated by recording the time the light pulse takes to travel from the LiDAR sensor to the ground, and back again. These measurements combined with other factors such as direction and location, allows users to work out the exact location of each pulse within the world. By emitting hundreds of thousands of these light pulses every second, the precise geometry of the earth below can be measured from these points of reflection. This is how a ‘point cloud’ is generated and the higher the number of points, the denser and more ‘realistic’ the 3D model is.
While large swathes of land can be captured by flying aircraft at great heights, point density tends to be reduced, producing results as low as 3-5 points per square metre. This is enough to create a DTM (Digital Terrain Model) and be used for something like measuring the height of trees in a forest, but the end result is basically two or more ‘surfaces’ which lack the kind of detail many clients require.
However, higher density data can be collected through Airborne Laser Scanning by flying lower with latest LiDAR technology. These screenshots on the right show a pointcloud collected by ROBIN +WINGS installed in a manned helicopter flown 100m above ground. Helicopters can fly low and slow, which means a greater number of pulses can be bounced back and forth collecting a greater number of measured points on the ground. For this particular project, point densities of 600pts/m2 were recorded, which meant the pointcloud was very detailed and precise. Think of a pointcloud as a very high-tech dot to dot puzzle, the fewer points in a dataset the more of a mystery the end result is!
Point density is not the only factor in producing such good data. It is important to minimise noise (outlying points which make the data look ‘messy’) and while a lot of this can be filtered out in post-processing, it is time consuming, so better to choose a high-precision sensor which does it automatically. Likewise, positional accuracy is also a big element. The data in this sample was accurate to 2cm, meaning every single point was within 2cm or less of its position in the real world. If point cloud data is not very accurate, it cannot be used to make measurements in comparison to the ‘real world’ (have a look at our helpful Laser Scanning Jargon guide for information about the technical terminology throughout this article).
In the data sample above, the features of interest are the high and medium voltage powerlines. At the 20-2000 point/m2 helicopter densities collected by ROBIN +WINGS, it is easy to measure the exact distances between the power cables, trees and roofs for safety and monitoring purposes. Many utility companies around the world are in process of switching from simple visual inspections to the use of LiDAR as in this example, thanks to the increased efficiencies and accuracies LiDAR offers. “Not only are accuracies increased while costs are reduced, but a modern LiDAR system such as ROBIN +WINGS can be mounted on a wide variety of aircraft including fixed wing planes and UAVs. Plus if 2000 pts/m2 is still not enough, the very same ROBIN system can be mounted on a backpack, to provide ultra-high densities of +10.000pts/m2, at 1-5cm accuracy” says our ALS expert Petri Nygrén.
For more information about Airborne Laser Scanning or to receive a copy of this data sample, please contact us.