YOLOv5-based Dense Small Target Detection Algorithm for Aerial Images Using DIOU-NMS

Simple item page
dc.contributor.authorWang, Yu
dc.contributor.authorZou, Xiang
dc.contributor.authorShi, Jiantong
dc.contributor.authorLiu, Minhua
dc.coverage.issue1cs
dc.coverage.volume33cs
dc.date.accessioned2024-05-28T12:43:34Z
dc.date.available2024-05-28T12:43:34Z
dc.date.issued2024-04cs
dc.description.abstractWith the advancement of various aerial platforms, there is an increasing abundance of aerial images captured in various environments. However, the detection of densely packed small objects within complex backgrounds remains a challenge. To address the task of detecting multiple small objects, a multi-object detection algorithm based on distance intersection over union loss non-maximum suppression (DIOU-NMS) integrated with you only look once version 5 (YOLOv5) is proposed. Leveraging the YOLOv5s model as the foundation, the algorithm specifically addresses the detection of abundantly and densely packed targets by incorporating a dedicated small object detection layer within the network architecture, thus effectively enhancing the detection capability for small targets using an additional upsampling operation. Moreover, conventional non-maximum suppression is replaced with DIOU-based non-maximum suppression to alleviate the issue of missed detections caused by target density. Experimental results demonstrate the effectiveness of the proposed method in significantly improving the detection performance of dense small targets in complex backgrounds.en
dc.formattextcs
dc.format.extent12-23cs
dc.format.mimetypeapplication/pdfen
dc.identifier.citationRadioengineering. 2024 vol. 33, iss. 1, s. 12-23. ISSN 1210-2512cs
dc.identifier.doi10.13164/re.2024.0012en
dc.identifier.issn1210-2512
dc.identifier.urihttps://hdl.handle.net/11012/245662
dc.language.isoencs
dc.publisherSpolečnost pro radioelektronické inženýrstvícs
dc.relation.ispartofRadioengineeringcs
dc.relation.urihttps://www.radioeng.cz/fulltexts/2024/24_01_0012_0023.pdfcs
dc.rightsCreative Commons Attribution 4.0 International licenseen
dc.rights.accessopenAccessen
dc.rights.urihttp://creativecommons.org/licenses/by/4.0/en
dc.subjectObject Detectionen
dc.subjectYOLOv5en
dc.subjectDIOU-NMSen
dc.subjectAerial Imagesen
dc.subjectSmall Object Detectionen
dc.subjectComplex Backgrounds.en
dc.titleYOLOv5-based Dense Small Target Detection Algorithm for Aerial Images Using DIOU-NMSen
dc.type.driverarticleen
dc.type.statusPeer-revieweden
dc.type.versionpublishedVersionen
eprints.affiliatedInstitution.facultyFakulta eletrotechniky a komunikačních technologiícs
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
24_01_0012_0023.pdf
Size:
1.21 MB
Format:
Adobe Portable Document Format
Description:
Download
Collections
2024/1