Ground Control Points (GCPs) are well-noticeable spots with known coordinates on the ground in an area of interest, measured on the center of the aerial targets/GCPs with a GNSS station/receiver. There are a multitude of GCP marking patterns, like cross marking, dot marking, chess-board marking patterns, etc. The basic of a good aerial target is something of highly contrasting color (black and white, other reflective colors) that is large enough to be seen from the air (i.e. 60cm x 60cm). The central point of the GCP should be clearly visible to a pixel level on the drone images. The GCPs need to be secured to the ground before starting the aerial survey with the drone both on the perimeter and scattered all over the target area equally. Make sure that the GCPs won’t move after the measurement and that the GCPs won’t be covered during drone flight.
Drone data are georeferenced thanks to the Global Positioning System (GPS) and other positioning sensors mounted on drones, meaning that each image contains information about the exact location in the space in their metadata. But the location can be subjected to errors or inaccuracies due to several potential atmospheric disturbances. That is where GCPs come in handy. These points with well-known accurate coordinates increase the absolute accuracy (both horizontal and vertical) of an aerial mapping project, placing the produced map or model from drone data at the exact geographical location on the Earth (with latitude-longitude-elevation or X-Y-Z coordinates). These points can then reduce the shift from the real location due to GPS from meters to centimeters accuracy. GCPs are essential to properly georeference drone deliverables, especially for specific applications where high accuracy is needed, such as orthomosaic maps and terrain models (DTMs, DSMs) used as input data for flood and hydraulic modelling or infrastructure surveys.
The area extent and topography will determine the general distribution of the GCPs for achieving effective and accurate results.
To plan the GCP layout you can use Google Earth Pro or Google My Maps, creating a new map and importing the KML downloaded from the mission in the Globhe platform. Using the satellite map as basemap you can add markers and start placing the GCPs following some recommendations:
Place GCPs around 300 meters apart, so a total of about 10 to 20 GCPs every 100 hectares should be placed starting from the corners of the flight area. An example of GCP layout is shown in the following picture.
After setting up GCPs the location can be measured using a survey-grade GNSS rover. GNSS rovers/receivers are the core product for satellite positioning. They convert signals from visible satellites into a geographical position on the Earth. The amount of visible satellites in the rover is dependent on the number of constellations the tool is compatible with, such as GPS, Glonass, Galileo, and Beidou.
Here are some of the most famous and recommended GNSS rovers:
To measure the accurate geographical location of the GCP, the GNSS station pole should be placed in the center of the target, maintaining it perfectly perpendicular to the ground (vertical position, as illustrated in the following image) and then follow the instructions in the receiver software (normally connected by smartphone or tablet). The target measurements can be performed before or after the actual drone aerial survey, as long as the aerial targets don’t move from their original location on the ground.
Guidelines for GCPs Measurements:
* Coordinate Reference System (CRS) = standardized system for defining and interpreting geographic coordinates, allowing for accurate and consistent positioning of features on the Earth's surface. In the context of drone data and mapping, the CRS plays a vital role in establishing a common reference point for geospatial data captured by the drone.
The collected coordinates for each aerial target will be saved in a table that can be downloaded in a .csv file. In the table, every row corresponds to one ground control point with all the information about latitude, longitude, elevation/altitude (sometimes X, Y, Z format, depending on the settings), and accuracy details (if available).
This table will be then used during data processing, in the workflow of photogrammetry, the process that overlaps photographs of an object, structure, or space, and converts them to produce drone data products, such as 2D maps or 3D digital models.