elib
DLR-Header
DLR-Logo -> http://www.dlr.de
DLR Portal Home | Imprint | Privacy Policy | Contact | Deutsch
Fontsize: [-] Text [+]

Increased Robustness of Object Detection on Aerial Image Datasets using Simulated Imagery

Konen, Kai and Hecking, Tobias (2021) Increased Robustness of Object Detection on Aerial Image Datasets using Simulated Imagery. In: 4th IEEE International Conference on Artificial Intelligence and Knowledge Engineering, AIKE 2021. 3rd IEEE Conference on Artificial Intelligence and Knowledge Engineering, 2021-12-01, Laguna Hills, CA. doi: 10.1109/AIKE52691.2021.00007. ISBN 978-1-6654-3736-3. (In Press)

[img] PDF - Only accessible within DLR
13MB

Abstract

Machine learning-based models for object detectionrely on large datasets of labeled images, such as COCO orImageNet. When models trained on these datasets are appliedto aerial images recorded on Unmanned Aerial Vehicles (UAVs),the problem arises that the conditions under which the trainingimages were created (for example, light, altitude, or angle) maybe different in the environment where the UAVs are put intopractice, leading to failed detections. This problem becomes evenmore pressing in safety critical applications where failures canhave huge negative impacts and also constitutes an obstaclefor certification of cognitive components in UAVs. Along a casestudy on car detection in low-altitude aerial imagery, we showthat using, both, artificial and real images for model traininghas a positive effect on the performance of object detectionalgorithms when the trained model is applied on images fromanother domain. Since simulated images are easy to create andobject labels are inherently given, the presented approach showsa promising direction for scenarios where adequate datasets aredifficult to obtain, as well as for targeted exploration of weakpoints of object detection algorithms.

Item URL in elib:https://elib.dlr.de/147100/
Document Type:Conference or Workshop Item (Speech)
Title:Increased Robustness of Object Detection on Aerial Image Datasets using Simulated Imagery
Authors:
AuthorsInstitution or Email of AuthorsAuthor's ORCID iDORCID Put Code
Konen, KaiUNSPECIFIEDhttps://orcid.org/0000-0001-7957-4477UNSPECIFIED
Hecking, TobiasUNSPECIFIEDhttps://orcid.org/0000-0003-0833-7989UNSPECIFIED
Date:December 2021
Journal or Publication Title:4th IEEE International Conference on Artificial Intelligence and Knowledge Engineering, AIKE 2021
Refereed publication:Yes
Open Access:No
Gold Open Access:No
In SCOPUS:Yes
In ISI Web of Science:Yes
DOI:10.1109/AIKE52691.2021.00007
ISBN:978-1-6654-3736-3
Status:In Press
Keywords:Object Detection, Aerial Image Datasets, Simu-lation, UAVs, YoloV4
Event Title:3rd IEEE Conference on Artificial Intelligence and Knowledge Engineering
Event Location:Laguna Hills, CA
Event Type:international Conference
Event Date:1 December 2021
HGF - Research field:Aeronautics, Space and Transport
HGF - Program:Aeronautics
HGF - Program Themes:other
DLR - Research area:Aeronautics
DLR - Program:L - no assignment
DLR - Research theme (Project):L - no assignment
Location: Köln-Porz
Institutes and Institutions:Institute for Software Technology
Institute for Software Technology > Intelligent and Distributed Systems
Deposited By: Hecking, Dr. Tobias
Deposited On:13 Dec 2021 10:36
Last Modified:24 Apr 2024 20:45

Repository Staff Only: item control page

Browse
Search
Help & Contact
Information
electronic library is running on EPrints 3.3.12
Website and database design: Copyright © German Aerospace Center (DLR). All rights reserved.