Generating a map using this technique can be a big problem, one of those problems is so critical with the consequences of missing valuable months of useful work when you do not have previous experience in this task.
The founders of Aerotas Mapping System We are told in an article by POB Online, that many surveyors focus on this work, first, discussing the type of drone they will acquire and then focus on discussing the characteristics of the final product they wish to obtain, resulting in the unnecessary extension of the time we discuss.
Faced with this situation, what is advisable, which leads to greater efficiency and profitability, is to start with the result to be obtained, identifying the sequence of work to be done and then implement the dron software to obtain the result.
We can, then, establish 3 steps to perform the work, namely, first make sure that the data collected in the field is reliable and correct; then, process this data in order to obtain an orthophoto and a digital elevation model (DEM); finally, using the created model, generate a surface in AutoCAD (or similar) as well as the 'line-work' (online work) and the final survey. Let's analyze the steps enunciated in detail:
Collect valid data in the field
In order for the teams to carry out a correct collection of information, it is required that the operators have been previously trained in the best practices that allow establishing both ground control and having an automatic pilot software configured to create topographic cartography.
In the case of the drone ground control adjustment, the same criteria used for conventional photogrammetry must be taken into account. The practice indicates that the objectives have been established and analyzed by surveying the terrain and its surroundings. The ideal is to establish five objectives per flight area, 4 at the corners and one at the center, being able to include more objectives according to the characteristics of the area. (high or low points).
Then, the autopilot is configured, taking into account the slightly overshoot of each control on both sides and capture two lines of photos beyond each control point using a graphical interface similar to that of Google Earth that allows tracing the area of the terrain and set the altitude of the flight.
Obtaining orthophoto and DEM
The second step is to process the photos taken by the drone to generate the orthophoto and the DEM. For this process you can choose among the multiple existing solutions in the market taking into account that the process follows the same logic as conventional photogrammetry. By this we mean that the photos are superimposed based on the shared land points through overlapping photos.
It should be noted that drones use smaller, uncalibrated cameras compared to those used in photogrammetry. So many photos must be taken to achieve a high overlap. This implies, for each point of the ground, an amount that oscillates between 9 and 16 photos, which through the technique of image recognition used by the chosen program will identify the 'mooring points' shared in the photos.
Removal of lifting surface and line work
It is in this last step that most surveying companies have greater difficulties due to the fact that the majority of modeling programs in 3D (such as Civil 3D) are not designed to work with the large surface models generated by the drone programs. That is why post processing solutions emerge as the right ones for this task.
Through these, the surveyor chooses the work points by clicking on those desired points in the digital image. Each of these is recorded by the program as a pair of coordinates.
Each point is, then, located in layers that coincide with the conventions established by Civil 3D (or the one you use) in such a way that when you open the file in that program, the points have a format similar to those from a GPS standard rover station or a total station.
Following this methodology of work can be achieved a dramatic saving of time and money in topographic mapping projects, estimating in an 80% savings over time. We can verify this by comparing the capture of points by conventional surveying done by an expert at 60 points per hour with the 60 points taken in a second by post-processing software.
Finally, always remember that the key to success and saving in working time is to identify the appropriate work sequence that will produce the desired result in the most efficient way possible.