计算机科学与探索 ›› 2026, Vol. 20 ›› Issue (6): 1795-1808.DOI: 10.3778/j.issn.1673-9418.2508062

• 图形图像 • 上一篇    下一篇

WMamba:基于Mamba和边缘感知增强的遥感弱目标检测

董燕1,2,王珍1,高广帅1+,高君宇3,李春雷1,刘洲峰1   

  1. 1. 中原工学院 信息与通信工程学院,郑州 450007
    2. 电子科技大学 自动化工程学院,成都 611731
    3. 西北工业大学 光电与智能研究院,西安 710072
    + 通信作者  E-mail: 6911@zut.edu.cn
  • 收稿日期:2025-08-25 修回日期:2025-10-24 在线发布日期:2026-06-01 出版日期:2026-05-29
  • 基金资助:
    国家自然科学基金(62301623,62472463,61873293);河南省重点研发项目(241111220700);中原工学院青年硕士生导师培育计划(SD202416);河南省国际科技合作一般项目(252102521049);河南省高等学校重点科研项目(25A620001);中原工学院自然科学基金重点项目(K2026ZD021);河南省自然科学基金青年基金B类项目(262300421225)。

WMamba:Remote Sensing Weak Target Detection Based on Mamba and Edge-Aware Enhancement

DONG Yan1,2, WANG Zhen1, GAO Guangshuai1+, GAO Junyu3, LI Chunlei1, LIU Zhoufeng1   

  1. 1. School of Information and Communication Engineering, Zhongyuan University of Technology, Zhengzhou 450007, China 
    2. School of Automation Engineering, University of Electronic Science and Technology, Chengdu 611731, China
    3. School of Artificial Intelligence, Optics and Electronics, Northwestern Polytechnical University,Xi'an 710072,China
    + Corresponding author  E-mail: 6911@zut.edu.cn
  • Received:2025-08-25 Revised:2025-10-24 Online:2026-06-01 Published:2026-05-29
  • Supported by:
    This work was supported by the National Natural Science Foundation of China (62301623, 62472463, 61873293), the Key Research and Development Project of Henan Province (241111220700), the Cultivation Program for Young Master Supervisor of Zhongyuan University of Technology (SD202416), the International Science and Technology Cooperation Program of Henan Province (General Category) (252102521049), the Key Scientific Research Project of Henan Higher Education Institutions (25A620001), and the Key Program of the Natural Science Foundation of Zhongyuan University of Technology (K2026ZD021), and the Youth Science Foundation Project (Category B) of the Natural Science Foundation of Henan Province (262300421225).

摘要: 针对遥感图像中因背景复杂、低对比度目标边界模糊、小目标像素占比低等导致的弱目标检测难的问题,提出了一种基于边缘感知增强的遥感弱目标检测算法WMamba。设计了一种耦合局部特征提取机制的局部Mamba模块,该模块通过将状态空间模型与局部感受野增强策略相结合,在保留长距离依赖建模优势的基础上,显著提升了对细节特征的捕获能力。该结构被用于替代主干网络及颈部网络中的C2f模块,通过逐层提取并融合多尺度上下文信息与局部纹理特征,逐步构建出语义层次丰富、判别性强的特征表示。为进一步强化目标轮廓信息,提出了一种边缘增强模块。该模块基于正负权重差异化分布策略,构建梯度级边缘信息提取路径,通过增强高频梯度响应并抑制低频背景噪声,有效凸显模糊边界和弱目标的轮廓特征,引导网络实现边缘区域的显著性表达。为了实现对增强特征中关键信息的精准聚焦,设计了多维注意力模块,从空间位置、通道关系及卷积核三个维度同时进行注意力权重分配。该机制能够有效抑制复杂背景干扰,强化目标区域响应,从而提高弱目标与背景的可区分性,提升检测精度。所提方法在NWPU VHR-10数据集以及UAVOD-10数据集上相较于基线模型YOLOv8n的mAP分别提升了5.5和3.2个百分点。

关键词: 遥感图像, 目标检测, Mamba, 上下文信息, 边缘增强

Abstract: To address the challenges of detecting weak targets in remote sensing images, such as complex backgrounds, low-contrast blurry target boundaries, and low pixel proportions of small targets, a weak target detection algorithm named weak Mambal (WMamba), based on edge-aware enhancement, is proposed. A local Mamba module coupled with a local feature extraction mechanism is designed. This module integrates state space models with a local receptive field enhancement strategy, preserving the advantage of long-range dependency modeling while significantly improving the ability to capture detailed features. This structure is used to replace the C2f modules in both the backbone and neck networks, enabling the progressive extraction and fusion of multi-scale contextual information and local texture features. This process gradually builds feature representations with rich semantic hierarchies and strong discriminative power. To further enhance target contour information, an edge enhancement module is introduced. This module employs a strategy of differentiated positive and negative weight distribution to construct a gradient-level edge information extraction path. By amplifying high-frequency gradient responses and suppressing low-frequency background noise, it effectively highlights the contour features of blurry boundaries and weak targets, guiding the network to achieve salient representations in edge regions. To enable precise focus on key information within the enhanced features, a multi-dimensional attention module is designed. This module allocates attention weights simultaneously across three dimensions: spatial position, channel relationships, and convolutional kernels. This mechanism effectively suppresses interference from complex backgrounds and strengthens responses in target regions, thereby improving the distinguishability between weak targets and the background and enhancing detection accuracy. The proposed method achieves mean average precision (mAP) improvements of 5.5 and 3.2 percentage points compared with the baseline model YOLOv8n on the NWPU VHR-10 and UAVOD-10 datasets, respectively.

Key words: remote sensing imagery, target detection, Mamba, contextual information, edge enhancement