《计算机应用》唯一官方网站 ›› 2026, Vol. 46 ›› Issue (9): 2968-2976.DOI: 10.11772/j.issn.1001-9081.2025080985

• 多媒体计算与计算机仿真 • 上一篇    

基于动态感知和交叉调制的遥感小目标检测

姚华(), 杨高明, 李雪莲, 陆凯旋   

  1. 安徽理工大学 计算机科学与工程学院,安徽 淮南 232001
  • 收稿日期:2025-08-28 修回日期:2025-10-10 接受日期:2025-10-16 发布日期:2026-09-16 出版日期:2026-09-10
  • 通讯作者: 姚华
  • 作者简介:姚华(2001—),男,安徽铜陵人,硕士研究生,CCF会员,主要研究方向:目标检测
    杨高明(1974—),男,安徽阜阳人,教授,博士,主要研究方向:计算机视觉、数字图像处理
    李雪莲(2005—),女,安徽淮南人,硕士研究生,主要研究方向:工业异常检测
    陆凯旋(2000—),男,安徽合肥人,硕士研究生,主要研究方向:目标检测。
  • 基金资助:
    国家自然科学基金资助项目(52374155);安徽省自然科学基金资助项目(2308085MF218)

Remote sensing small object detection with dynamic perception and cross-modulation

Hua YAO(), Gaoming YANG, Xuelian LI, Kaixuan LU   

  1. School of Computer Science and Engineering,Anhui University of Science and Technology,Huainan Anhui 232001,China
  • Received:2025-08-28 Revised:2025-10-10 Accepted:2025-10-16 Online:2026-09-16 Published:2026-09-10
  • Contact: Hua YAO
  • About author:YAO Hua,born in 2001, M. S. candidate. His research interestsinclude object detection.
    YANG Gaoming, born in 1974, Ph. D., professor. His researchinterests include computer vision, digital image processing.
    LI Xuelian,born in 2005, M. S. candidate. Her research interestsinclude industrial anomaly detection.
    LU Kaixuan, born in 2000, M. S. candidate. His research interestsinclude object detection.

摘要:

针对遥感图像中小目标检测背景干扰严重以及多尺度特征提取和融合不充分导致检测精度下降的问题,提出一种动态感知和交叉调制网络(DPCMNet)。首先,为增强模型对细粒度目标的感知能力,设计多层次的混合特征聚合(HFA)模块,以动态强化小目标区域的响应强度,有效地丰富纹理和结构特征;其次,提出动态调制融合(DMF)模块,引入通道和空间联合调制机制,实现特征间的自适应交互,从而缓解多尺度融合过程中的信息丢失,并提升小目标的定位和识别能力;最后,基于DMF模块构建动态调制融合金字塔(DMFP),通过高层语义信息和浅层细节补偿的跨尺度关联,实现特征的渐进式融合与增强。实验结果显示,DPCMNet在USOD(Unicorn Small Object Dataset)和VEDAI(VEhicle Detection in Aerial Imagery)遥感小目标数据集上均提升了检测性能,该网络的平均精度(AP)较基准模型YOLOv11n分别提高了2.9和3.1个百分点。

关键词: 遥感图像, 小目标检测, 上下文信息增强, 特征交互, 多尺度融合

Abstract:

To address the challenges of severe background interference and insufficient multi-scale feature extraction and fusion for small object detection in remote sensing images, which lead to degraded detection accuracy, a Dynamically Perceptive and Cross-Modulated Network (DPCMNet) was proposed. First, to enhance the model’s perception ability to fine-grained objects, the multi-level Hybrid Feature Aggregation (HFA) module was designed to strengthen the response intensity of small object regions dynamically, thereby enriching the textural and structural features effectively. Second, the Dynamic Modulation Fusion (DMF) module was proposed to introduce a joint channel and spatial modulation mechanism, so as to enable adaptive feature interaction, thereby alleviating information loss during multi-scale feature fusion and improving localization and recognition capabilities for small objects. Finally, based on the DMF module, the Dynamic Modulation Fusion Pyramid (DMFP) was constructed to achieve progressive feature fusion and enhancement through cross-scale association between high-level semantic information and low-level detail compensation. Experimental results demonstrate that DPCMNet achieves significant performance improvements on the Unicorn Small Object Dataset (USOD) and VEhicle Detection in Aerial Imagery (VEDAI) remote sensing small object datasets, with Average Precision (AP) increased by 2.9 and 3.1 percentage points, respectively, compared to the baseline model YOLOv11n.

Key words: remote sensing image, small object detection, context information enhancement, feature interaction, multi-scale fusion

中图分类号: