Control of unmanned aerial vehicles based on the detection algorithm

The work analyzes the current tasks of unmanned aircraft, which showed the need to develop algorithms and systems for automatic tracking of objects operating in real-time mode, in the tasks of conducting reconnaissance flights and targeting precision weapons. An algorithm for detecting objects in the image by correlation filtration based on a two-dimensional discrete Fourier transform for control problems of unmanned aerial vehicles is developed. A software implementation of the detection algorithm using the OpenCV / C ++ computer vision library (EmguCV / C #) is obtained. A number of methods of computer libraries used for constructing program code are presented. Time schedules and recommendations for using the algorithm in different shooting conditions are given.