Gümüşhane ÜniversitesiGümüşhane ÜniversitesiAçık Erişim Arşivi · RDL Open

Saw-YOLOv5: Scale-Aware YOLOv5 for Object Detection in Aerial Images

BildiriAçık Erişim

Özet

The detection of objects in aerial images is impor-tant for many real world problems related to military defense, transportation, and etc. However, this is a challenging task as a result of the presence of various scales of objects in the same image, the large variety of contexts across aerial images, various brightness levels due to image acquisition at different times of the day and so on. To address these challenges, this paper introduces Saw-YOLOv5 for object detection in aerial images. Saw-YOLOv5 is a deep network based on YOLOv5, which was proposed for object detection in natural images. Saw-YOLOv5 extends YOLOv5 with the addition of several attention modules in its design. The results of our experiments, conducted on the aerial dataset delivered by the Turkey Technology Team for the Artificial Intelligence in Transportation Competition, showed that Saw-YOLOv5 outperforms previous methods, particularly for pedestrian detection, by yielding a mean mAP of 80.23% over all objects.

Anahtar Kelimeler

aerial image analysisattention mechanismdeep networksObject detection
RDL Kalıcı Bağlantı (PID)
https://rdlopen.rawdatalibrary.net/r/gumushane/12718Resmi atıflarda kullanılan kalıcı adres
İndeks kayıtları
ScopusScopus
Kaydın indeks veri tabanlarındaki sayfası (aktarımda alınan kayıt numarasıyla). Erişim, indeksin kendi koşullarına bağlıdır.
Atıflar
Scopus
1
OpenAlex
1
Her kaynağın kendi saydığı toplam atıf; aktarım ve zenginleştirme sırasında alındı.
Tam Kayıt
AlanDeğerDil
dc.contributor.authorDoruk, Abdullah Enes
dc.contributor.authorAlgul, Mucteba
dc.contributor.authorAkyurek, Feyzullah
dc.contributor.authorAlpaydm, Osman Kursat
dc.contributor.authorUslu, Fatmatulzehra
dc.date.accessioned2026-10-22T10:20:27.154730867Z
dc.date.available2026-10-22T10:20:27.154730867Z
dc.date.issued2023
dc.description.abstractThe detection of objects in aerial images is impor-tant for many real world problems related to military defense, transportation, and etc. However, this is a challenging task as a result of the presence of various scales of objects in the same image, the large variety of contexts across aerial images, various brightness levels due to image acquisition at different times of the day and so on. To address these challenges, this paper introduces Saw-YOLOv5 for object detection in aerial images. Saw-YOLOv5 is a deep network based on YOLOv5, which was proposed for object detection in natural images. Saw-YOLOv5 extends YOLOv5 with the addition of several attention modules in its design. The results of our experiments, conducted on the aerial dataset delivered by the Turkey Technology Team for the Artificial Intelligence in Transportation Competition, showed that Saw-YOLOv5 outperforms previous methods, particularly for pedestrian detection, by yielding a mean mAP of 80.23% over all objects.
dc.identifier.doi10.1109/hora58378.2023.10155778
dc.identifier.isbn979-835033752-5
dc.identifier.scopus2-s2.0-85165688341
dc.identifier.urihttps://rdlopen.rawdatalibrary.net/r/gumushane/12718
dc.language.isoen
dc.publisherInstitute of Electrical and Electronics Engineers Inc.
dc.relation.conference5th International Congress on Human-Computer Interaction, Optimization and Robotic Applications, HORA 2023
dc.relation.ispartofHORA 2023 - 2023 5th International Congress on Human-Computer Interaction, Optimization and Robotic Applications, Proceedings
dc.rightsinfo:eu-repo/semantics/openAccess
dc.subjectaerial image analysis
dc.subjectattention mechanism
dc.subjectdeep networks
dc.subjectObject detection
dc.subject.otherAntennas
dc.subject.otherMilitary applications
dc.subject.otherMilitary photography
dc.subject.otherObject recognition
dc.subject.otherAerial image analysis
dc.subject.otherAerial images
dc.subject.otherAttention mechanisms
dc.subject.otherDeep network
dc.subject.otherMilitary defense
dc.subject.otherNatural images
dc.subject.otherNetwork-based
dc.subject.otherObjects detection
dc.subject.otherPedestrian detection
dc.subject.otherReal-world problem
dc.subject.otherObject detection
dc.titleSaw-YOLOv5: Scale-Aware YOLOv5 for Object Detection in Aerial Images
dc.typeConference Object
rdl.citation.openalex1
rdl.citation.scopus1
rdl.conference.date8 June 2023 through 10 June 2023
rdl.conference.placeIstanbul
rdl.oa.statusgreen
rdl.oa.urlhttps://hdl.handle.net/20.500.12885/6552
rdl.source.indexSCOPUS