SPDD-YOLO for Small Object Detection in UAV Images

Simple item page
dc.contributor.authorXue, Xiangen_US
dc.contributor.authorJi, Ya Tuen_US
dc.contributor.authorLiu, Yangen_US
dc.contributor.authorXu, H. T.en_US
dc.contributor.authorRen, Q. D. E. J.en_US
dc.contributor.authorShi, B.en_US
dc.contributor.authorWu, N. E.en_US
dc.contributor.authorLu, M.en_US
dc.contributor.authorZhuang, X. F.en_US
dc.contributor.editorChen, Renjieen_US
dc.contributor.editorRitschel, Tobiasen_US
dc.contributor.editorWhiting, Emilyen_US
dc.date.accessioned2024-10-13T18:06:18Z
dc.date.available2024-10-13T18:06:18Z
dc.date.issued2024
dc.description.abstractAerial images captured by drones often suffer from blurriness and low resolution, which is particularly problematic for small targets. In such scenarios, the YOLO object detection algorithm tends to confuse or misidentify targets like bicycles and tricycles due to the complex features and local similarities. To address these issues, this paper proposes a SPDD-YOLO model based on YOLOv8. Firstly, the model enhances its ability to extract local features of small targets by introducing the Spatial-to- Depth Module (SPDM). Secondly, addressing the issue that SPDM reduces the receptive field, leading the model to overly focus on local features, we introduced Deep Separable Dilated Convolution (DSDC), which expands the receptive field while reducing parameters and forms the Deep Dilated Module (DDM) together with SPDM. Experiments on the VisDrone2019 dataset demonstrate that the proposed model improved precision, recall, and mAP50 by 5.8%, 5.7%, and 6.4%, respectively.en_US
dc.description.sectionheadersPosters
dc.description.seriesinformationPacific Graphics Conference Papers and Posters
dc.identifier.doi10.2312/pg.20241327
dc.identifier.isbn978-3-03868-250-9
dc.identifier.pages2 pages
dc.identifier.urihttps://doi.org/10.2312/pg.20241327
dc.identifier.urihttps://diglib.eg.org/handle/10.2312/pg20241327
dc.publisherThe Eurographics Associationen_US
dc.rightsAttribution 4.0 International License
dc.rights.urihttps://creativecommons.org/licenses/by/4.0/
dc.subjectCCS Concepts: Computing methodologies → Object recognition; Object identification
dc.subjectComputing methodologies → Object recognition
dc.subjectObject identification
dc.titleSPDD-YOLO for Small Object Detection in UAV Imagesen_US
Files
Original bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
pg20241327.pdf
Size:
2.8 MB
Format:
Adobe Portable Document Format
Download
Collections
PG2024 Conference Papers and Posters