YOLOv11n-Ghost-P2: An Accuracy–Efficiency Study of Lightweight Small-Vehicle Detection in Aerial Images
Applied Sciences · 09.11.2026 · 16 (21) · 11095
DOI: 10.3390/app162111095 ↗Kurz gefasst
Fachartikel in der Zeitschrift Applied Sciences, online erschienen am 09.11.2026. Verfasst von Liang Wu und 2 weiteren Autor:innen. Die Arbeit nennt 30 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Hubei Provincial Department of Education, National Natural Science Foundation of China.
Abstract (Originalsprache)
Small-vehicle detection in aerial imagery requires fine spatial detail within limited computational budgets. We propose YOLOv11n-Ghost-P2, which combines Ghost-based feature compression with a high-resolution P2 fusion branch. Relative to standard YOLOv11n, the model reduces parameters by 38.5%, floating-point operations (FLOPs) by 39.7%, and single-patch latency by 47.9%. On the source-scene-disjoint validation partition of the Dataset for Object Detection in Aerial Images (DOTA), mean average precision (mAP@0.5) is 0.6222 for YOLOv11n and 0.5367 for the proposed model, while strict mAP@0.5:0.95 is 0.3545 and 0.2953, respectively. On an NVIDIA Tesla T4 platform, the model processes 37.7 patches/s; mean full-scene time across three large images decreases from 2.39 to 1.26 s. A five-seed ablation comparing the dual-Ghost architecture with and without P2 shows the largest AP@0.5 improvement for Tiny vehicles, with a mean gain of 0.01662. On five additional scenes, tiny-vehicle recall increases from 21/34 to 26/34 with P2. The results characterize the accuracy–efficiency trade-off of allocating part of the compression savings to high-resolution detection.
Förderung
- Hubei Provincial Department of Education
- National Natural Science Foundation of China
Zitieren
APA
Wu, L., Tu, P., & Li, X. (2026). YOLOv11n-Ghost-P2: An Accuracy–Efficiency Study of Lightweight Small-Vehicle Detection in Aerial Images. Applied Sciences, 16(21), 11095. https://doi.org/10.3390/app162111095
Vancouver
Wu L, Tu P, Li X. YOLOv11n-Ghost-P2: An Accuracy–Efficiency Study of Lightweight Small-Vehicle Detection in Aerial Images. Applied Sciences. 2026;16(21):11095. doi:10.3390/app162111095
BibTeX
@article{Wu2026,
author = {Wu, Liang and Tu, Pengcheng and Li, Xiang},
title = {YOLOv11n-Ghost-P2: An Accuracy–Efficiency Study of Lightweight Small-Vehicle Detection in Aerial Images},
journal = {Applied Sciences},
year = {2026},
volume = {16},
number = {21},
pages = {11095},
doi = {10.3390/app162111095}
}
EUVIMED