Open AccessDetection and Tracking of UAV Targets Using Deep Learning
Author(s) -
Mohamed Khedir Noraldain Alamin
Publication year - 2021
Language(s) - English
DOI - 10.54388/jkues.v1i2.51
Subject(s) - drone , clutter , computer science , radar cross section , tracking (education) , radar , artificial intelligence , process (computing) , scope (computer science) , deep learning , aeronautics , real time computing , tracking system , interference (communication) , computer vision , engineering , telecommunications , channel (broadcasting) , psychology , pedagogy , genetics , kalman filter , biology , programming language , operating system
In recent years, the use of Flying drones and modern Unmanned aerial vehicles (UAVs) with the latest techniques and capabilities for both civilian and military applications growing sustainably on a large scope, Drones could autonomously fly in several environments and locations and could perform various missions, providing a system for UAV detection and tracking represent crucial importance. This paper discusses Designing Detection and Tracking method as a part of Aero-vehicle Defense System (ADS) for UAVs using Deep learning algorithms. The small Radar cross-section (RCS) foot-print makes a problem for Traditional methods and Aero-vehicle Defense systems to distinguish between birds, stealth fighters, and UAVs incomparable of size and RCS characteristics, the detection is a challenge in low RCS targets because the chance of detection is incredibly less moreover, in the existence of interference and clutter which reduce the performance of detection process rapidly.