3.2.2. UAV Data Collection Methods

To prove the concept of UAV data capture as a remote sensing technology for land rights mapping in Kenya, an exploratory research investigation was undertaken. This included the entire UAV-based workflow, starting from the choice of UAV equipment, pilot and flight training, flight authorization, and the final data collection in the field which was carried out in two different sites in Kajiado County: a rural area in Mailua and the township of Kajiado. To accommodate the different characteristics of the flight locations, two different UAVs were chosen (see Figure 7) both with RGB sensors on board.

**Figure 7.** UAV data collection with the DT18 in Mailua and the Phantom 4 in Kajiado.

In Mailua, the DT18, a fixed-wing UAV with a long endurance and a large range was selected. In contrast, the vertical take-off and landing UAV DJI Phantom 4 was the preferred equipment to capture data of Kajiado, as the urban area did not provide large spaces for take-off and landing. Both study sites, were captured with indirect georeferencing (Figure 8), i.e., Ground Control Points (GCPs) were distributed within the field and measured with a Global Navigation Satellite System (GNSS) achieving a final accuracy of less than 2 cm. RGB orthomosaics and digital surface models (DSM) of approximately 6 cm Ground Sample Distance (GSD) were generated with Pix4DMapper. Three tiles of 300 × 300 m were selected to demonstrate the boundary mapping approach.

**Figure 8.** Areas of investigation of 300 × 300 m and a 6 cm GSD of Kajiado, Kenya.

The evaluation of the UAV workflow was based on the case study results from Kenya, as well as statistics of the UAV image processing, and resulted in a SWOT analysis. Further insights were gained from a stakeholder assessment of the potential of UAV-based technology to capture land rights in Kenya [47].
