Scholarlink

<p>Detecting small, overlapping objects in high-resolution remote sensing imagery is crucial for applications such as smart city monitoring and disaster response. However, challenges such as severe feature confusion and spatial misalignment hinder accurate localization. This paper introduces Multiscale SOG-DETR, a systematic redesign of the RT-DETR framework tailored for remote sensing small-object detection. We propose a lightweight Multiscale Overlapping-Object Decoupling Network (MOODNet) to significantly reduce feature entanglement in overlapping regions. Additionally, our specialized fusion neck, comprising the Residual Spatial-Alignment Progressive Fusion Module (SAPFM), E-CGAFusion, and WTConv2d modules, enhances multiscale semantic focus and preserves high-frequency details cost-effectively. Experimental results on the RSOD, VisDrone2019, and NWPU VHR-10 datasets demonstrate that Multiscale SOG-DETR achieves superior detection accuracy with significantly fewer parameters compared to the baseline RT-DETR model, increasing <InlineEquation ID="IEq1"> <EquationSource Format="TEX">\(AP_{IoU=50}\)</EquationSource> </InlineEquation> by 3.1%, 3.0%, and 5.2%, and <InlineEquation ID="IEq2"> <EquationSource Format="TEX">\(AP_{IoU=50:95}\)</EquationSource> </InlineEquation> by 5.1%, 2.1%, and 8.5%, respectively. These findings position Multiscale SOG-DETR as a promising solution for efficient and accurate small-object detection in remote sensing applications. The source code is publicly available at <a href="https://github.com/AaronWang-code/Multiscale-SOG-DETR">https://github.com/AaronWang-code/Multiscale-SOG-DETR</a>.</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

Multiscale feature optimization for accurate small object detection in remote sensing imagery

  • 期刊论文

  • 发表日期：

    2026年6月25日

查看全文
  • Bingxiang Wang,
  • Mugen Zhou,
  • Wenzhuo Ma,
  • Tianyu Li,
  • Changsheng Zhu

摘要

Detecting small, overlapping objects in high-resolution remote sensing imagery is crucial for applications such as smart city monitoring and disaster response. However, challenges such as severe feature confusion and spatial misalignment hinder accurate localization. This paper introduces Multiscale SOG-DETR, a systematic redesign of the RT-DETR framework tailored for remote sensing small-object detection. We propose a lightweight Multiscale Overlapping-Object Decoupling Network (MOODNet) to significantly reduce feature entanglement in overlapping regions. Additionally, our specialized fusion neck, comprising the Residual Spatial-Alignment Progressive Fusion Module (SAPFM), E-CGAFusion, and WTConv2d modules, enhances multiscale semantic focus and preserves high-frequency details cost-effectively. Experimental results on the RSOD, VisDrone2019, and NWPU VHR-10 datasets demonstrate that Multiscale SOG-DETR achieves superior detection accuracy with significantly fewer parameters compared to the baseline RT-DETR model, increasing \(AP_{IoU=50}\) by 3.1%, 3.0%, and 5.2%, and \(AP_{IoU=50:95}\) by 5.1%, 2.1%, and 8.5%, respectively. These findings position Multiscale SOG-DETR as a promising solution for efficient and accurate small-object detection in remote sensing applications. The source code is publicly available at https://github.com/AaronWang-code/Multiscale-SOG-DETR.