GCPs for Lidar

Quantity and Distribution

Just like photogrammetry, lidar data needs GCPs in order to be properly georeferenced to the real world. Since RTK streaming is used in the collection of the lidar data, we have pretty high confidence in the relative accuracy of each swath of lidar data, or each flight path the drone takes. Because of the RTK streaming, we don’t need as many GCPs as you would in a traditional non-RTK photogrammetry mission. Instead, you want to make sure you have an absolute bare minimum of 3 GCPs for each flight, offset from one another (not in a straight line) but we recommend 5 as an added redundancy for reliability. 

Quality and Appearance

In photogrammetry, we like to see GCPs that are large enough to be visible, easily located in photos, and have a precise center point so there is no question as to where the measurement was taken. 

Lidar GCPs are no different, however, the way lidar sees IS different than your typical RGB sensor so there are some considerations to take into account. Most lidar sensors can either display their data with an RGB coloring or at least a reflectivity gradient on the points as it collects the data meaning your typically colored GCPs will probably work. If you don’t have RGB, reflectivity is the next best thing to think about. Contrasting colors typically have different reflective properties, so if you set a checkerboard GCP tile, you should be able to see it in the lidar data. To be safe, black and white, or black and another bright color will probably show up best in a reflectivity gradient.

Try to avoid 3D GCPs. While it’s true that raised or 3D objects show up well in lidar data, measuring them and setting them can lead to a cumbersome and less reliable workflow. It’s much more straightforward to rely on differentiating reflective properties to make your GCPs pop. This way, you don’t have to modify your field GCP workflow much at all from what you might be used to with photogrammetry.

Again, keep it simple!