Before one starts acquiring images, certain parameters need to be determine and set in the mission planner application (Hernandez-Lopez et al., 2013; Mangiameli et al., 2013; Mesas-Carrascosa et al., 2015). Figure 5.7 exemplifies a mission planner interface (eMotion) that allows to plan flights. It is possible to prepare this in office before heading to the field. The parameters include: identification and specification of take-off and landing sites, flying height, direction and speed, number of flight lines, amount of overlaps. These parameters have an influence on the acquired images and the flight campaign. It is, therefore, important to appreciate what goes into each of these parameters and what deserves attention when setting them.
- Flying height: one needs to determine how high to fly. This depends on the required spatial resolution of the images and the focal length of the camera in use. Naturally, an inverse relationship exists between spatial resolution and flying height. For the Canon S110 NIR, for example, a flying height of 145 m will result in a spatial resolution of 5 cm. On the other hand, a flying height of about 290 m will produce 10 cm spatial resolution. If detailed information is needed that captures, for instance, crop canopy characteristics, then a lower flying height is needed: 35 m elevation will give approximately 1 cm resolution. Note, however, that the image footprint covered reduces with lower flying height. In other words, a larger extent with a flying height of 290 m is covered than when flying at 145 m (assuming the same camera). It is, therefore important to consider this critically before selecting. See more details here.
- Overlaps: to allow for the generation of a height model (e.g., digital surface or canopy model), images must be acquired with a decent overlap. As the name suggests, a flight path is the 3D line that the UAV follows in acquiring images. Typical flight paths are often space-filling curves with long and mutually parallel flight lines (legs) that optimize the time available for image acquisition, and thus minimize the time lost to making turns, or overflying terrain for which we have no interest. The overlap between two successive images in direction of the flight path is termed forward or longitudinal overlap, while the overlap between two images acquired on adjacent legs of the flight path is known as lateral overlap. Longitudinal and lateral overlaps ensure that a common area is imaged on two successive or adjacent images. This is needed to generate stereo images, because with these a height model can be generated. The amount of forward or lateral overlap specified influences the number of legs as well as the number of images per leg. Larger overlaps will lead to more legs and more images per leg. One typically plans for image overlaps between 70 and 80%. These overlaps are recommended if one needs accurate height models.
- Take-off and landing: take-off and landing sites must be defined as waypoints either inside or outside the mission area. For large sites, it may be important to have the take-off and landing sites located in the middle. Fixed-wing UAVs need some form of runway, though the eBee proved quite forgiving, and octocopter models can normally get by with just a few square metres. Ideally, these sites must be devoid of obstacles such as trees, tree stumps, powerlines, large rocks and boulders. In heavy agricultural areas, rows of crops such as groundnuts may be considered as take-off and landing sites. With fixed-wings, it is important to consider wind direction, because take-off and landing must be with head-wind. This enables it to fly more stable in low speed stages. See here for more details.
- Wind speed: Wind speed and direction can be specified in a good mission planning application (e.g., eMotion for eBee). Providing this information allows a better estimation of the time required to fly an area.