Drone Services - Surveying
LiDAR vs Photogrammetry
When deploying drones for surveying purposes – we have two core technologies at our disposal – LiDAR and Photogrammetry. Which one is better suitable for your project greatly
depends on the purpose of the survey and the final deliverables and their use. Both technologies have their strengths and weaknesses. Sometimes is beneficial to use a combination of
both. First we need to understand their main characteristics and how they work.

LiDAR
LiDAR stands for “Light Detection And Ranging”. It uses laser (as the source of “light”) to determine ranges by targeting an object or a surface and measuring the time for the reflected light to return to the receiver. Most modern LiDAR systems use multiple “returns”, allowing them to “see through” vegetation and detect the internal structure of the vegetation and even the ground below. This is a key difference from Photogrammetry, which can only detect the top surface (first reflection). As the drone flies over the survey area – the LiDAR generates dense 3D point cloud based on these returns. The density of the point cloud depends on the laser frequency, the height of the drone above the terrain and the forward speed of the drone. The deployed lasers are usually monochromatic, which results in single-colour point cloud. However, most modern LiDAR systems include RGB camera – which allows for each individual point to be assigned its respective true RGB colour. The point cloud can also be colorised in real-time based on variety of characteristics – such as overall height / altitude, type or order of the individual laser return, or the reflectance value of the target.

Photogrammetry
Photogrammetry uses normal RGB camera to take series of images covering the survey area. The actual capture pattern and image overlap is really important, as each point on the ground must be captured in multiple images from different angles. After the survey flight is completed – complex algorithms based on trigonometry are then used to process the data and calculate the coordinates of each point to generate the 3D point cloud. This process is extremely computing heavy and requires powerful computers and significant amount of time. This is in strike contrast to LiDAR, which generates the 3D point cloud in realtime during the flight without the need for any data processing after the flight. Also because of the complex detection algorithms needed to identify the same point in all the captured images – the resulting 3D point cloud is not as detailed or accurate as the one generated by the LiDAR.
Both 3D point clouds need to then be “post-processed” to clean them up and generate the desired deliverables, such as DTM (Digital Terrain Model), DSM (Digital Surface Model), DEM (Digital Elevation Model), Orthomosaic (precise undistorted 2D image of the survey area as seen from above in direct orthographic projection – it can also include the elevation data for each pixel) and others…
The key advantage of LiDAR is the workflow speed, accuracy and no need to process the data after the flight to generate the 3D point cloud. Main disadvantage is the cost, as
LiDAR systems have traditionally been much more expensive than basic RGB cameras. However, the cost of these systems is gradually coming down and makes them more
competitive with the RGB camera solutions. Photogrammetry is generally more affordable and easier to deploy, but the cost can easily build up during the necessary processing of
the RGB imagery after the survey flight in order to generate the base 3D point cloud.

Geolocation / GCP’s
In order to “geolocate” the 3D point cloud in the real world coordinate system – we deploy GCP’s = Ground Control Points. These are strategically positioned across the survey area prior to the drone flights and their coordinates are measured with survey-grade professional equipment with millimetre-level accuracy. Once the 3D point cloud is generated – the GCP’s are used to precisely align the whole point cloud to the realworld coordinates.
Once properly geolocated – the point cloud (along with all derivative outputs, such as DEM, DTM, DSM, Orthomosaic, etc…) can be used for precise measurements, construction planning, property line determination and many other survey / post-survey tasks that require real-world dimensions / coordinates.
Deliverables


The base deliverable for property survey is the geolocated 3D point cloud. We can deliver this in a format compatible with your SW suite for further post-processing on your end.

When desired – we have the capability to derivate other useful data from the initial 3D point cloud. The most common request are two base DEM’s (Digital Elevation Models) – DSM & DTM. DSM (Digital Surface Model) represents the property as it is – including vegetation, buildings and other structures. DTM (Digital Terrain Model) represents the base terrain (ground) level with vegetation, building and other structures removed.

Another common request for additional deliverables is the Orthomosaic. This is a top-down undistorted
orthographic view of your property where each pixel corresponds to real-world coordinates. Orthomosaic,
because of its accuracy, is commonly used as the source for cartography – generating various types of
property maps.

Orthomosaic is also commonly used to extract or map various features of your property, such as property
boundary, specific features (roads, buildings, etc…). It is also commonly used for construction planning and
as a base for architectural drawings. Please contact us with your specific requirements.