Tracking of Unmanned Aerial Vehicles Using Computer Vision Methods: A Comparative Analysis

Stanisław Hożyń,Miłosz Wierszyło

doi:10.2478/sjpna-2020-0014

Stanisław Hożyń, Miłosz Wierszyło

Open Access

https://doi.org/10.2478/sjpna-2020-0014

Copy DOI

Abstract

Abstract Tracking of small objects in any given airspace is an integral part of modern security systems. In these systems, there are embedded methods that employ the techniques based on either radio waves, or acoustic signals, or light radiation. The computer vision operation, springing from the light radiation-based technique, has prompted interest in its research. This operation has the advantage of being less expensive than radars and acoustic systems. In addition, it can solve complex security problems by detecting and tracking humans, vehicles, and flying objects. Therefore, this article evaluates the usefulness of the varying computer vision algorithms for tracking of small flying objects.

Highlights

The trackers were tested on video sequences, in which the movement of Unmanned aerial vehicles’ (UAVs) differed in respect of speed, dynamics, and distance from the camera
Analysis of the above algorithms has shown that the CSRT algorithm is the best choice for UAVs tracking
The MOSSE method, which was very fast, achieved seventy-six per cent efficacy. This efficacy is too low for tracking applications, but it can be improved using image pre-processing steps, cooperation with detection algorithm, or the Kalman and Particle filters

Summary

Introduction

Acoustic systems analyse sound emitted by the object in order to detect the direction of signal emission They do not require seeing the object during operation, which allows to use them in urban facilities, or behind hills [4]. In this respect, acoustic systems outperform other methods based on radio waves and visible light. A large number of microphones, and advanced digital signal processing enable determination of azimuth, and elevation of the target in real-time. Another advantage involves modularity and scalability, which allow the use of this type of detection in combination with other systems [6]. Their highly complex solution systems have the ability to detect, recognise, and track various targets, such as humans, vehicles, or flying objects in any surrounding area [12]

Objectives

Methods

Results

Conclusion