Overview At present, one of the most commonly used technique for topographic surveys is aerial photogrammetry. This technique uses aerial images to determine the geometric properties of objects and spatial situations. Therefore, three-dimensional information is obtained from two-dimensional images. It is important to keep in mind that drones are only the platforms where the sensors that collect the information are installed (in this case cameras to capture images), and that depending on the type of work do not have to be the best option. If it is determined that the use of a UAS platform is the best option, it must be taken into account that there is a great variety of UASs in the market, and it is necessary to choose the most appropriate system depending on the needs of the project. Due to the great advance in drone technology, have emerged some UAS systems for photogrammetry that offer a very simple workflow of one-button survey. However, it is necessary to take into account the importance of having knowledge about photogrammetry in order to use this technique, since leaving all procedures and decision making in the hands of a system can be risky. Low cost systems have the advantage that the initial investment to acquire these equipments is lower compared to professional systems. However, they are usually closed systems that do not allow modifications to adapt them to the different types of jobs that may arise. In this case, the biggest limitation of this type of equipment is that the camera is integrated into the system, so you could not use another camera that can be better adapted to the needs. Objectives The aim of this article is to compare different UAS systems for use in aerial photogrammetry. As mentioned, there is a wide variety of equipments available in the market. In this case we have tried to compare low cost systems that offer complete and easy solutions with professional systems specifically designed for aerial photogrammetry. The ultimate goal is to compare the performance that different equipment offers us in a photogrammetric flight.
1 Materials and Methods 1.1 Materials To perform the tests, two multirotor of different characteristics have been used. In this case the Phantom 4 model was chosen as an example of a UAS not specifically designed for aerial photogrammetry and the LOOKOUT VTOL model as an example of a UAS specifically designed for photogrammetry. Phantom 4 As an example of a UAS that offers an integrated and lower cost solution Phantom 4 of DJI. Figure 1: DJI Phantom 4 LOOKOUT VTOL As an example of a professional UAS designed specifically for aerial photogrammetry, the LOOK- OUT VTOL product of Throttle Aerospace Systems has been used. The main difference between the two systems is that the LOOKOUT VTOL model allows the integration of different sensors, while the Phantom 4 model has its own integrated RGB camera, so it is not possible to integrate any other sensor. Figure 2: Throttle Aerospace Systems LOOKOUT VTOL
Table 1: The following table summarizes the main features of both devices. Parameters Phantom 4 Lookout VTOL Endurance (min) 20 40 Payload in (g) 250 2000 Physical size in (mm) 350 350 500 500 175 Range in (km) 2 8 As you can see, endurance and payload capacity are major difference between both systems, more endurance helps to airborne the sensor. Table 2: Comparison table of the sensors. Parameters Phantom 4 Sensor Lookout VTOL Sensor Sensor Size (mm) 6.20 4.65 23.5 15.6 Resolution (píxel) 4000 3000 (12.4 Mp) 6000 4000 (24.3 Mp) Pixel size (um) 1.55 4.2 Focal length (mm) 3,6125 16 ISO sensitivity 100-1.6000 100-16.000 Shutter speed (s) 8-1/8.000 30-1/4000 1.2 Methods To make the comparison between both systems, two missions have been planned, one with each of the equipment and with the sensors mentioned above. One of the most important factors when planning a photogrammetric flight is the GSD (Ground Sample Distance), since it is the spatial resolution (the detail) of the images that will be obtained from the aerial flights, that is to say, the distance that corresponds to each pixel in the terrain. The GSD need to a job depends on the needs of this. As the objective of these tests has been to compare systems with different characteristics, flight heights have been adapted to the terrain (H) to obtain a similar GSD with both systems and to make comparable comparisons. Table 3: Comparison table of the relationship between flight height and GSD obtained with the different systems. To obtain a similar GSD with both systems, the flight of the Phantom 4 has been planned at a height of 60 meters (GSD: 2.57 cm) and the LOOKOUT VTOL flight at a height of 100 meters (GSD: 2.65 cm). On the other hand, the missions have been planned with the same speed of the aircraft and obtaining the same overlaps between images, so that both flights were comparable.
Height (m) GSD of Phantom 4 Sensor (cm) GSD of Lookout VTOL (cm) 60 2.57 1.59 80 3.43 2.12 120 5.15 3.18 In this case the values used for these parameters were as follows: Aircraft speed: 6 m/s Longitudinal overlap: 70% Transversal overlap: 70% In both cases, planning has been done trying to cover the largest possible area on each flight. For this, the limitation of the flight time of each aircraft has been taken into account. The following table shows the most determinant parameters used in each of the flights. Table 4: Comparison table of plannings Parameters Phantom 4 Lookout VTOL Overlaps 70% / 70% 70% / 70% Flight height (m) 60 100 GSD (cm) 2.57 2.65 Flight time (min) 25 40 2 Results Using the mentioned flight parameters, the missions were planned and the following results were obtained related to the performance of the two systems: Table 5: Comparison table of results. Parameters Phantom 4 Sensor Lookout VTOL with Sony Alpha 6000 No of images 260 (Aprox) 450 (Aprox) Output Image Size (Mb) 12 Mb 10 Mb Dimensions of each photography (m) 103 77.2 158.75 105.8 Area of each photography (Acre) 1.96 3.45 Total covered area (Acre) 42 166
As you can see in this table, the results obtained in terms of performance are much better using the professional LOOKOUT VTOL system. In a single flight it is possible to cover a three times greater area using the LOOKOUT VTOL in comparison with the Phantom 4. Therefore, it would be necessary to make three flights with the Phantom 4 to cover the same area that LOOKOUT VTOL (with the Sony Alpha 6000 camera) is able to cover in a single flight, considering that in both cases we want to obtain the same spatial resolution (GSD). This would mean an increase of time, since to realize three flights with the Phantom 4 would be necessary 75 minutes, against the 45 of the LOOKOUT VTOL. And that without taking into account that between different flights it is necessary to land, take off, change battery, check the new flight plan. On the other hand, the volume of photographs obtained to cover the same area would be higher with the Phantom 4, since in three flights it would be necessary to make 1027 (12.5 Gb data) photographs in front of the 450(4.5 Gb data) that would be necessary with the LOOKOUT VTOL to cover the same area with the same resolution. 3 Conclusions As mentioned, there are significant differences between the two systems in terms of performance. In this aspect, LOOKOUT VTOL wins. The saving of flight time (field work) and image processing (office work), saves time when doing a project, which translates into a saving of money. In addition, LOOKOUT VTOL allows the integration of different sensors, offering the possibility of using other sensors that can further increase this difference in performance. Not to mention that it is possible to integrate other cameras like multispectral or thermal. Therefore, low-cost drones can be an economical option for small jobs less than 1sqkm projects, but for larger jobs or in which different sensors are need to adapt the equipment to different projects, it is necessary to go to a professional system.