Scholarlink

<p>This study proposes an improved Residual Asymptotic Feature Pyramid Network (R-AFPN) to address challenges in small target detection from the Unmanned Aerial Vehicle (UAV) perspectives, such as scale imbalance, feature extraction difficulty, occlusion, and computational constraints. The R-AFPN integrates three key modules: Residual Asymptotic Feature Fusion (RAFF) for adaptive spatial fusion and cross-scale linking, Shallow Information Extraction (SIE) for capturing detailed shallow features, and Hierarchical Feature Fusion (HFF) for bottom-up incremental fusion to enhance deep feature details. Experimental results demonstrate that R-AFPN-L achieves 50.7% AP<sub>50</sub> on the TinyPerson dataset and 48.9% mAP<sub>50</sub> on the VisDrone2019 dataset, outperforming the baseline by 3% and 1.2%, respectively, while reducing parameters by 15.1%. This approach offers a lightweight, efficient solution for small target detection in UAV applications.</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

R-AFPN: a residual asymptotic feature pyramid network for UAV aerial photography of small targets

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年5月9日

查看全文
  • Zuowen Chen,
  • Yahong Ma,
  • Zi’an Gong,
  • Minghao Cao,
  • Yuyao Yang,
  • Zhiyuan Wang,
  • Tengjie Wang,
  • Jing Li,
  • Yuxi Liu

摘要

This study proposes an improved Residual Asymptotic Feature Pyramid Network (R-AFPN) to address challenges in small target detection from the Unmanned Aerial Vehicle (UAV) perspectives, such as scale imbalance, feature extraction difficulty, occlusion, and computational constraints. The R-AFPN integrates three key modules: Residual Asymptotic Feature Fusion (RAFF) for adaptive spatial fusion and cross-scale linking, Shallow Information Extraction (SIE) for capturing detailed shallow features, and Hierarchical Feature Fusion (HFF) for bottom-up incremental fusion to enhance deep feature details. Experimental results demonstrate that R-AFPN-L achieves 50.7% AP50 on the TinyPerson dataset and 48.9% mAP50 on the VisDrone2019 dataset, outperforming the baseline by 3% and 1.2%, respectively, while reducing parameters by 15.1%. This approach offers a lightweight, efficient solution for small target detection in UAV applications.