Note: The text on this page will load first, there is a high quality image that will load after the text. It may take up to 30 seconds depending on your internet connection.
The technology commercially available for unmanned aerial vehicles (UAV), commonly referred to as drones, is changing at a dramatic rate. Now, in early 2017, high quality UAVs are available with accurate GPS and professional grade cameras.
One of the most common uses of UAV technology is map making. Light weight yet high quality cameras carried by drones can take unprecedented quality aerial photos. There are many uses for these aerial photos from industrial inspection, crop plant health, mining volume calculations, aerial photography and map making. Drones equipped with GPS allow each photo to be tagged with the exact position the drone was in when the picture was taken. By taking multiple overlapping photos, software programs are able to stitch together these photos to create 2 dimensional maps and 3 dimensional models.
In the map below, approximately 380 photos were taken by a drone and combined with software to produce a map of a housing complex in Indonesia. Let’s go through the process of how it was done.
First the boundaries of the area to be mapped must be created. Google Maps satellite images and location data are often the tool of choice for creating these boundaries. The satellite images in Google Maps can be several years old, are low precision however it is a good starting point.
Once the boundary of the area to be mapped has been defined, the height of the drone flight needs to be determined. Generally speaking, the closer to the ground the more detail can be seen in each individual photo. However more photos are required to map the entire area in the boundary. The more photos involved, the larger the data set involved and the more computing power required to process all the images. In the Indonesian housing complex map above, the total data set exceeded 2 gigabytes.
The image below from The Swiss Foundation for Mine Action (FSD) shows a drone flying it’s pre-programmed flight path to cover the boundary area. The images must overlap on the side and front in order for the mapping software to stitch them together.
Image Credit: The Swiss Foundation for Mine Action (FSD)
Post processing of the image generated by mapping software may be required in order to fully complete the project. The Indonesian housing complex map shown has been rotated and cropped to produce the final result shown. Image rotation leaves corners of the image that are not squared off, and the photos can be enhanced through the use of filters and brightness or contrast modifications.