Efficient vision transformers with edge enhancement for robust small target detection in drone-based remote sensing

Small object detection in UAV remote sensing imagery faces significant challenges due to scale variations, background clutter, and real-time processing requirements. This study proposes a lightweight transformer-based detector, MLD-DETR, which enhances detection performance in complex scenarios thro...

Full description

Saved in:
Bibliographic Details
Main Authors: Xuguang Zhu, Zhizhao Zhang
Format: Article
Language:English
Published: Frontiers Media S.A. 2025-07-01
Series:Frontiers in Remote Sensing
Subjects:
Online Access:https://www.frontiersin.org/articles/10.3389/frsen.2025.1599099/full
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850096162670903296
author Xuguang Zhu
Zhizhao Zhang
author_facet Xuguang Zhu
Zhizhao Zhang
author_sort Xuguang Zhu
collection DOAJ
description Small object detection in UAV remote sensing imagery faces significant challenges due to scale variations, background clutter, and real-time processing requirements. This study proposes a lightweight transformer-based detector, MLD-DETR, which enhances detection performance in complex scenarios through multi-scale edge enhancement and hierarchical attention mechanisms. First, a Multi-Scale Edge Enhancement Fusion (MSEEF) module is designed, integrating adaptive pooling and edge-aware convolution to preserve target boundary details while enabling cross-scale feature interaction. Second, a Layered Attention Fusion (LAF) mechanism is developed, leveraging spatial depth-wise convolution and omnidirectional kernel feature fusion to improve hierarchical localization capability for densely occluded targets. Furthermore, a Dynamic Positional Encoding (DPE) module replaces traditional fixed positional embeddings, enhancing spatial perception accuracy under complex geometric perspectives through learnable spatial adapters. Combined with an Inner Generalized Intersection-over-Union (Inner-GIoU) loss function to optimize bounding box geometric consistency, MLD-DETR achieves 36.7% AP50% and 14.5% APs on the VisDrone2019 dataset, outperforming the baseline RT-DETR by 3.2% and 1.8% in accuracy while achieving 20% parameter reduction and maintaining computational efficiency suitable for UAV platforms equipped with modern edge computing hardware. Experimental results demonstrate the algorithm’s superior performance in UAV remote sensing applications such as crop disease monitoring and traffic congestion detection, offering an efficient solution for real-time edge-device deployment.
format Article
id doaj-art-708e7bc4ffdc4cd88f1d562eb032b108
institution DOAJ
issn 2673-6187
language English
publishDate 2025-07-01
publisher Frontiers Media S.A.
record_format Article
series Frontiers in Remote Sensing
spelling doaj-art-708e7bc4ffdc4cd88f1d562eb032b1082025-08-20T02:41:17ZengFrontiers Media S.A.Frontiers in Remote Sensing2673-61872025-07-01610.3389/frsen.2025.15990991599099Efficient vision transformers with edge enhancement for robust small target detection in drone-based remote sensingXuguang Zhu0Zhizhao Zhang1College of Innovation and Practice, Liaoning Technical University, Fuxin, ChinaSchool of Software, Liaoning Technical University, Huludao, ChinaSmall object detection in UAV remote sensing imagery faces significant challenges due to scale variations, background clutter, and real-time processing requirements. This study proposes a lightweight transformer-based detector, MLD-DETR, which enhances detection performance in complex scenarios through multi-scale edge enhancement and hierarchical attention mechanisms. First, a Multi-Scale Edge Enhancement Fusion (MSEEF) module is designed, integrating adaptive pooling and edge-aware convolution to preserve target boundary details while enabling cross-scale feature interaction. Second, a Layered Attention Fusion (LAF) mechanism is developed, leveraging spatial depth-wise convolution and omnidirectional kernel feature fusion to improve hierarchical localization capability for densely occluded targets. Furthermore, a Dynamic Positional Encoding (DPE) module replaces traditional fixed positional embeddings, enhancing spatial perception accuracy under complex geometric perspectives through learnable spatial adapters. Combined with an Inner Generalized Intersection-over-Union (Inner-GIoU) loss function to optimize bounding box geometric consistency, MLD-DETR achieves 36.7% AP50% and 14.5% APs on the VisDrone2019 dataset, outperforming the baseline RT-DETR by 3.2% and 1.8% in accuracy while achieving 20% parameter reduction and maintaining computational efficiency suitable for UAV platforms equipped with modern edge computing hardware. Experimental results demonstrate the algorithm’s superior performance in UAV remote sensing applications such as crop disease monitoring and traffic congestion detection, offering an efficient solution for real-time edge-device deployment.https://www.frontiersin.org/articles/10.3389/frsen.2025.1599099/fullUAVdrone-based remote sensingRT-DETRsmall object detectionmulti-scale edge enhancement
spellingShingle Xuguang Zhu
Zhizhao Zhang
Efficient vision transformers with edge enhancement for robust small target detection in drone-based remote sensing
Frontiers in Remote Sensing
UAV
drone-based remote sensing
RT-DETR
small object detection
multi-scale edge enhancement
title Efficient vision transformers with edge enhancement for robust small target detection in drone-based remote sensing
title_full Efficient vision transformers with edge enhancement for robust small target detection in drone-based remote sensing
title_fullStr Efficient vision transformers with edge enhancement for robust small target detection in drone-based remote sensing
title_full_unstemmed Efficient vision transformers with edge enhancement for robust small target detection in drone-based remote sensing
title_short Efficient vision transformers with edge enhancement for robust small target detection in drone-based remote sensing
title_sort efficient vision transformers with edge enhancement for robust small target detection in drone based remote sensing
topic UAV
drone-based remote sensing
RT-DETR
small object detection
multi-scale edge enhancement
url https://www.frontiersin.org/articles/10.3389/frsen.2025.1599099/full
work_keys_str_mv AT xuguangzhu efficientvisiontransformerswithedgeenhancementforrobustsmalltargetdetectionindronebasedremotesensing
AT zhizhaozhang efficientvisiontransformerswithedgeenhancementforrobustsmalltargetdetectionindronebasedremotesensing