The SKIF drone is an important component of Culver Aviation's system for collecting, storing and analyzing data on the state of the earth's surface and objects on it.
The imagery it captures allows Culver Aviation's GIS team to create orthophotomaps of high accuracy and quality and, using analytical tools, to produce clear and comprehensive reports for a wide range of services.
How SKIF drone helps to assess the consequences of destruction, design houses and find crop diseases, the head of geoinformation systems department of Culver Aviation Andriy Levchuk told to the ITC.ua.
The first generation of guided drones appeared during the World War. The loudest American project, the Kettering Bug, could drop its wings on command from the ground and hit its target - a kind of an ancestor of the cruise missile and the kamikaze drone.
Now drones are much smarter and more functional. They not only perform complex military tasks, but also help to increase crops and assess damage from missile attacks - such as the Ukrainian drone SKIF.
The predecessor of our SKIF was a reconnaissance drone that was being developed in 2014. It had a video camera and a thermal imager. This equipment was enough to find enemy units at any time of day. No more was required of this drone.
In 2016, the military aircraft was upgraded for civilian tasks - that's how the SKIF model appeared. The video camera and thermal imager were replaced with a powerful camera with a 61-megapixel sensor and a full-frame 35-mm sensor.
With the new "eyes", the aircraft can shoot the surface of the Earth with a resolution of up to 1 cm per pixel. This is enough to see cracks in the asphalt or distinguish the structure of plants from a height of 100 meters.
SKIF is now part of Culver Aviation's geographic information system. It is a complex for collecting, storing and analyzing data about the condition of the Earth's surface and objects on it.
There is more information in the images SKIF takes than can be seen at a glance. Analytical tools and a little bit of experience make it possible to get a lot of value out of an ordinary crisp photo. Here are a few case studies.
Measure the field area and find problem areas
Imagine you have an area of land of several thousand hectares and you want an image of it. In other words, a huge photo of the highest quality, showing any object referenced to a given coordinate system.
This is what is called an orthophoto. Using special software, it is stitched from the photos taken by the drone during the flight. With the orthophotoplane you can calculate the exact area of the site, the surface area of areas with height differences, etc.
The orthophotoplane is a basic service for farmers. On it, we can determine the clear boundaries of the areas allocated to crops. And when the crop grows, we can see problems in the field: sowings, places where the ladder has not risen, soaked or trampled areas.
The agronomist still has to go to the field himself, but the preliminary analysis of the orthophoto makes the inspection not chaotic, but orderly.
Here's an example. The photo above is from SKIF. It shows areas where the crop has been affected by one factor or another. You can't see this if you only examine the edges of the field, and satellite photos won't give that much detail or relevance.
Also, high-resolution images are used to check different objects - from the quality of the ground surface to the condition of roads, pipelines and any extended objects.
Create a volumetric model for a construction project
Any picture taken by the Skif can be digitized and a digital terrain model can be obtained. How SKIF, a Ukrainian drone, helps assess the consequences of destruction, design houses and find diseases of crops. The outer surfaces of objects in such a model consist of clouds of a huge number of points in a three-dimensional coordinate system.
DEM is the simplest and most reliable way to get information about terrain relief. Not a single BIM project How Ukrainian SKIF drone helps to assess the consequences of devastation, design houses and find crop diseases is implemented without highly accurate information about the local landscape.
While orthophotomaps are used by farmers and cartographers, point cloud analytics is more attracted to building and utility designers.
For example, when designing an apartment complex, you can calculate the best places to install surveillance cameras. Or you can see where rainwater will accumulate and plan a square with moisture-loving plants.
With DMM it is possible to calculate with 90% accuracy the amount of ground objects - materials on a construction site or rock in a quarry.
Detect crop diseases in time
Multispectral imagery, that is imagery in specific ranges of the electromagnetic spectrum, is used in the restoration of ancient manuscripts, detection of fakes, and weather forecasting.
The main advantage of such imagery is that it allows you to see the nuances that are inaccessible to human vision.
For agrarians, multispectral imaging helps:
- determine the degree of plant vegetation;
- determine the chlorophyll content of plants;
- make yield forecasts;
- identify the source of the problem at early stages and solve it quickly.
Detail is also important for multispectral imaging. Plant diseases start with small foci that can only be seen in clear photographs, taken and processed in time. In large areas, only unmanned aircraft such as SKIF, complete with the work of GIS specialists, can cope with such a task.
Evaluate the effects of destruction
New techniques for analyzing UAV imagery involve artificial intelligence and machine learning. Together with location information, these technologies have been used for several years to calculate damage from earthquakes, hurricanes, tsunamis, tornadoes, and fires.
Skifs can be used to estimate the consequences of floods and other natural disasters such as forest fires and hurricanes. In Ukraine, damage from missile and artillery attacks added to the list of disasters. Information about them can be used to calculate material damage and compensation from the aggressor.
To create a system that can do this level of analytics, you need a neural network and an array of data. And the more data, the better the machine will learn to distinguish damage and classify it.
With this system, GIS specialists analyze visual information from the drone and output it as a report.
A recent example - Culver Aviation specialists flew over one of the villages in Chernihiv region, which was damaged during the Russian occupation, and identified the damaged buildings.
To order service or to investigate on your own?
An orthophotoplane can also be made using a copter, but the latter is only suitable for monitoring small areas. In one day, a copter can fly over several hundred hectares, depending on the model and the number of batteries. At this speed, the customer will not be able to quickly get the necessary analytics.
But aircraft-type UAVs can easily cope with voluminous tasks. SKIF can cover up to 1,500 hectares per battery charge. If a company needs to quickly shoot tens of thousands of hectares, a larger fleet can be involved. Culver Aviation, for example, has one of the largest fleets in Ukraine for just such cases.
How quickly the customer receives an analytical report depends not only on the capabilities of the drone, but also on the level of computing power. For example, according to the average flight data of the Skif on a single battery, an orthophoto taken at a resolution of 1 cm per pixel can weigh up to 70 GB.
To work with large agricultural holdings or energy providers, home capacity is not enough. The same goes for the skills to prepare analytical reports. That is why our company changed its business model of drone sales to Drone-as-a-Service several years ago. With this system, the customer does not spend money on maintaining a fleet of aircraft, specialists collecting and analyzing information, and redundant server capacity.
The range of services provided by drones is growing: aerial photography combined with modern technology covers business needs and helps save time, manpower and, in the end, money.