Real-Time Object Detection and Classification by UAV Equipped With SAR.

Krzysztof Gromada,Wojciech Stecz,Barbara Siemiątkowska,Krystian Płochocki,Karol Woźniak

doi:10.3390/s22052068

Abstract

The article presents real-time object detection and classification methods by unmanned aerial vehicles (UAVs) equipped with a synthetic aperture radar (SAR). Two algorithms have been extensively tested: classic image analysis and convolutional neural networks (YOLOv5). The research resulted in a new method that combines YOLOv5 with post-processing using classic image analysis. It is shown that the new system improves both the classification accuracy and the location of the identified object. The algorithms were implemented and tested on a mobile platform installed on a military-class UAV as the primary unit for online image analysis. The usage of objective low-computational complexity detection algorithms on SAR scans can reduce the size of the scans sent to the ground control station.

Highlights

Unmanned aerial vehicles are the fastest-growing segment of the military aerospace market
This article deals with synthetic aperture radar (SAR) that can be mounted on platforms with a maximum take-off weight (MTOW) of up to 150 kg
The paper presents a new concept of classification methods: YOLOv5 is used to classify objects and determine the areas in which objects are located

Summary

Introduction

Unmanned aerial vehicles are the fastest-growing segment of the military aerospace market. Their sensors’ capabilities form the critical feature of each unit type. A wide range of UAV payloads (including light electro-optical/infrared (EO/IR) systems, synthetic aperture radars (SAR) [1], SIGINT (signal intelligence), and EW (electronic warfare)) is systematically enriched and introduced to the military market. This technological trend is connected with civilian demand to use unmanned aerial systems in search-and-rescue missions or environmental protection tasks. Another important application of SAR radars is UAV navigation support in all weather conditions [3–5]

Objectives

Methods

Results

Conclusion