航空学报 > 2025, Vol. 46 Issue (23): 631694-631694   doi: 10.7527/S1000-6893.2025.31694

干扰环境下无人机多源感知专栏

基于类关系挖掘的遥感图像广义小样本分割方法

丛润民1,2, 孙豪言1,2, 罗宇轩3(), 方豪1,2   

  1. 1.山东大学 控制科学与工程学院,济南 250061
    2.机器智能与系统控制教育部重点实验室,济南 250061
    3.香港城市大学 电脑科学系,香港 999077
  • 收稿日期:2024-12-19 修回日期:2025-03-04 接受日期:2025-04-23 出版日期:2025-05-13 发布日期:2025-05-06
  • 通讯作者: 罗宇轩 E-mail:yuxuanluo4-c@my.cityu.edu.hk
  • 基金资助:
    山东省泰山学者青年专家项目(tsqn202306079);国家自然科学基金面上项目(62471278)

Generalized few-shot segmentation for remote sensing image based on class relation mining

Runmin CONG1,2, Haoyan SUN1,2, Yuxuan LUO3(), Hao FANG1,2   

  1. 1.School of Control Science and Engineering,Shandong University,Jinan 250061,China
    2.Key Laboratory of Machine Intelligence and System Control,Ministry of Education,Jinan 250061,China
    3.Department of Computer Science,City University of Hong Kong,Hong Kong 999077,China
  • Received:2024-12-19 Revised:2025-03-04 Accepted:2025-04-23 Online:2025-05-13 Published:2025-05-06
  • Contact: Yuxuan LUO E-mail:yuxuanluo4-c@my.cityu.edu.hk
  • Supported by:
    Taishan Scholar Project of Shandong Province(tsqn202306079);National Natural Science Foundation of China(62471278)

摘要:

随着遥感卫星和无人机技术的快速发展,遥感图像的获取变得更加方便,实现图像内容的精准解析逐渐成为研究热点。然而,现有遥感图像语义分割方法在仅有少量样本的情况下难以应对场景变化后出现的新类别目标,进而影响模型的整体性能。为应对遥感场景中可能出现的新类和新场景问题,引入了广义小样本分割技术,旨在利用少量样本快速适应新类,并高效完成遥感图像的语义分割任务。首先,针对遥感图像中不同类别和不同尺寸物体难以分辨的挑战,引入多尺度特征融合解码器,提高了模型对不同尺度目标的分割能力。然后,针对广义小样本分割任务中的背景语义偏移问题,设计了背景一致性建模训练策略,保证了模型对背景特征的准确表达。此外,设计了类间关系挖掘技术,利用大量基类数据生成的原型,增强少量数据生成的新类原型,使其能够从基类中获取有利信息,从而实现更精确的分割。最后,在公开数据集上,将本方法与主流方法进行对比实验,分割指标mIoU最高提升11.48%,证明了所提算法的有效性。

关键词: 遥感图像, 语义分割, 广义小样本, 知识蒸馏, 关系挖掘

Abstract:

With the rapid development of remote sensing satellites and unmanned aerial vehicle technologies, the acquisition of remote sensing images has become more convenient, making precise interpretation of image content a research hotspot. However, existing remote sensing image semantic segmentation methods struggle to handle new object categories that emerge due to scene changes when only a small number of samples are available, thereby affecting the overall performance of the model. To address the potential issues of new classes and new scenes in remote sensing scenarios, this paper introduces the generalized few-shot segmentation technique, aiming to rapidly adapt to new classes using a few samples and efficiently complete the semantic segmentation task for remote sensing images. Firstly, to tackle the challenge of distinguishing objects of different categories and different sizes in remote sensing images, a multi-scale feature fusion decoder is introduced to enhance the model’s segmentation ability for targets of different scales. Then, to address the background semantic shift problem in the generalized few-shot segmentation task, a background consistency modeling training strategy is designed to ensure the accurate expression of background features by the model. Furthermore, this paper proposes an inter-class relationship mining technique, which utilizes prototypes generated from a large amount of base class data to enhance the prototypes generated from a small amount of new class data, enabling them to acquire beneficial information from the base classes and achieve more precise segmentation. Finally, comparative experiments with mainstream methods are conducted on public datasets, and the mIoU segmentation metric is improved by up to 11.48%, demonstrating the effectiveness of the proposed algorithm.

Key words: remote sensing, semantic segmentation, generalized few-shot, knowledge distillation, relationship mining

中图分类号: