测绘通报 ›› 2024, Vol. 0 ›› Issue (10): 77-83.doi: 10.13474/j.cnki.11-2246.2024.1013.

• 学术研究 • 上一篇    

遥感图像农田识别的跨类别小样本分割方法

王星, 倪欢   

  1. 南京信息工程大学遥感与测绘工程学院, 江苏 南京 210044
  • 收稿日期:2024-01-03 修回日期:2024-06-03 发布日期:2024-11-02
  • 通讯作者: 倪欢,E-mail:nih@nuist.edu.cn
  • 作者简介:王星(1999—),男,硕士生,主要研究方向为小样本语义分割在遥感图像处理中的应用。E-mail:1152890400@qq.com
  • 基金资助:
    先进光学遥感技术北京市重点实验室开放基金(AORS202310)

Cross-category few-shot segmentation for farmland recognition in remote sensing images

WANG Xing, NI Huan   

  1. School of Remote Sensing & Geomatics Engineering, Nanjing University of Information Science & Technology, Nanjing 210044, China
  • Received:2024-01-03 Revised:2024-06-03 Published:2024-11-02

摘要: 深度学习驱动的遥感图像语义分割方法依赖大量人工标记样本,且对未知任务泛化能力差,尤其是在类别体系不断更新的精细化语义分割任务中,未知类别(训练样本中不存在的类别)的识别精度亟待提升。基于此,本文面向多类农田类别,提出了一种跨类别小样本分割方法。该方法设计了一种双分支结构,即支持分支与查询分支,其中支持分支用于分割先验的提取,查询分支用于完成分割先验的传播,获取查询图像的分割结果。此外,该方法应用查询特征生成自支持查询原型,显著提高了原型的表达能力;在支持集和查询集之间引入原型对齐正则化机制,充分利用来自支持集的知识,提升了分割器的判别能力。试验同时引入高空间分辨率和高光谱图像土地覆盖数据集,对本文方法性能进行了充分验证。试验结果表明,相较于现有小样本分割方法,本文方法在小样本条件下能够获得更为优异的跨类别农田识别结果。

关键词: 遥感图像, 语义分割, 小样本学习, 原型学习

Abstract: Deep learning-driven semantic segmentation methods for remote sensing images rely heavily on a large number of manually labeled samples and exhibit poor generalization for unknown tasks, especially in the fine-grained semantic segmentation task where the category system is constantly updated, and the recognition accuracy of the unknown categories (the categories that don't exist in the training samples) needs to be urgently improved. Based on this, the paper proposes a cross-category few-shot segmentation method aimed at multiple farmland categories. The method designs a dual-branch structure, comprising a support branch and a query branch, where the support branch is used for the extraction of segmentation prior, and the query branch is used to complete the propagation of segmentation prior and obtain the segmentation results of the query image. Additionally, the method applies query features to generate self-supporting query prototypes, which significantly improves the expressive ability of the prototypes; a regularization mechanism for prototype alignment between the support and query set is introduced, which makes full use of the knowledge from the support set and improves the discriminative ability of the segmentation. The experiments simultaneously introduce high spatial resolution and hyperspectral image land cover datasets to fully validate the performance of the proposed method. The experimental results show that compared with the existing few-shot segmentation methods, the proposed method can obtain more excellent cross-category farmland recognition results under few-shot conditions.

Key words: remote sensing images, semantic segmentation, few-shot learning, prototype learning

中图分类号: