Scholarlink

<p>In Unmanned Aerial Vehicle (UAV) aerial and remote sensing images with complex backgrounds, small object detection faces challenges such as local feature ambiguity, multi-scale semantic mismatch, and weak adaptability to geometric deformation. Therefore, we propose an Adaptive Multi-Scale Fusion Network (<b>AMFNet</b>), which enhances feature representation of small objects from three aspects: feature enhancement, cross-scale fusion, and deformation modeling. Firstly, the Spatial Structure Enhancement Module (<b>SSEM</b>) captures global dependencies using single-head self-attention, while heterogeneous convolution extracts local details, thereby achieving complementary global–local feature modeling. Secondly, we construct a Dual-guided Cross-scale Enhancement Path (<b>DCEP</b>), which aligns and complements multi-scale features through deep semantic guidance and shallow structural feedback, together with a cascaded loop fusion strategy. Finally, an A2-Dynamic Adaptive Module (<b>A2-DyAM</b>) is introduced, adapting to geometric deformation using deformable convolution, and a Position Adaptive Transformation (PAT) mechanism refines the features to further improve geometric accuracy. Experiments on the VisDrone2019 dataset demonstrate that AMFNet improves mAP50 by 12.0% over the baseline YOLOv12 and 12.7% over YOLOv26. These results indicate that the proposed method maintains stable detection performance under low-discriminative features and geometric variations.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

AMFNet: adaptive multi-scale fusion network for small object detection

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年7月4日

查看全文
  • Bao Tian,
  • MengNan Hu,
  • LingLing Li,
  • XueZhuan Zhao,
  • MengMeng Tang,
  • ZiYang Wang

摘要

In Unmanned Aerial Vehicle (UAV) aerial and remote sensing images with complex backgrounds, small object detection faces challenges such as local feature ambiguity, multi-scale semantic mismatch, and weak adaptability to geometric deformation. Therefore, we propose an Adaptive Multi-Scale Fusion Network (AMFNet), which enhances feature representation of small objects from three aspects: feature enhancement, cross-scale fusion, and deformation modeling. Firstly, the Spatial Structure Enhancement Module (SSEM) captures global dependencies using single-head self-attention, while heterogeneous convolution extracts local details, thereby achieving complementary global–local feature modeling. Secondly, we construct a Dual-guided Cross-scale Enhancement Path (DCEP), which aligns and complements multi-scale features through deep semantic guidance and shallow structural feedback, together with a cascaded loop fusion strategy. Finally, an A2-Dynamic Adaptive Module (A2-DyAM) is introduced, adapting to geometric deformation using deformable convolution, and a Position Adaptive Transformation (PAT) mechanism refines the features to further improve geometric accuracy. Experiments on the VisDrone2019 dataset demonstrate that AMFNet improves mAP50 by 12.0% over the baseline YOLOv12 and 12.7% over YOLOv26. These results indicate that the proposed method maintains stable detection performance under low-discriminative features and geometric variations.