测绘学报 ›› 2023, Vol. 52 ›› Issue (11): 1941-1952.doi: 10.11947/j.AGCS.2023.20220410

• 摄影测量学与遥感 • 上一篇    下一篇

高光谱影像少样例分类的无监督元学习方法

高奎亮1, 刘冰1, 余岸竹1, 徐佰祺2, 胡伟3, 胡家玮3   

  1. 1. 信息工程大学, 河南 郑州 450001;
    2. 31016部队, 北京 100089;
    3. 中国人民武装警察部队甘肃省总队, 甘肃 兰州 730046
  • 收稿日期:2022-06-27 修回日期:2023-02-08 发布日期:2023-12-15
  • 通讯作者: 刘冰 E-mail:liubing220524@126.com
  • 作者简介:高奎亮(1996-),男,博士生,主要研究方向为深度学习与遥感影像处理。E-mail:gokling1219@163.com
  • 基金资助:
    河南省自然科学基金(222300420387)

An unsupervised meta learning method for hyperspectral images few-shot classification

GAO Kuiliang1, LIU Bing1, YU Anzhu1, XU Baiqi2, HU Wei3, HU Jiawei3   

  1. 1. Information Engineering University, Zhengzhou 450001, China;
    2. Troops 31016, Beijing 100089, China;
    3. Gansu Provincial Corps of the Chinese People's Armed Police Force, Lanzhou 730046, China
  • Received:2022-06-27 Revised:2023-02-08 Published:2023-12-15
  • Supported by:
    The Natural Science Foundation of Henan Province (No. 222300420387)

摘要: 针对高光谱影像少样例分类问题,本文提出了一种无监督元学习方法。该方法能够利用无标记样本进行无监督元学习,在显著减少深度学习模型对大量标记样本依赖的同时有效提高高光谱影像少样例分类精度。首先,基于自监督学习思想利用主成分分析和数据增强方法构造同一样本的不同增强特征,以形成大量不同的任务并对模型进行元训练。然后,分别利用目标影像中随机选择的少量标记样本和全部样本对模型进行微调和分类测试,评估模型在少样例条件下的分类性能,并在微调和分类过程中引入投票策略进一步提高分类精度。本文方法能够在零标记样本的条件下对模型进行充分的无监督元训练,突破了监督元学习方法需要大量源标记样本的瓶颈和限制。3组公开数据集上的试验结果表明,本文方法在少样例条件下能够获得较现有方法更为优异的分类结果。

关键词: 高光谱影像分类, 少样例分类, 无监督元学习, 深度学习, 投票策略

Abstract: Aiming at the problem of hyperspectral images (HSIs) few-shot classification, an unsupervised meta learning method is proposed in this paper. This method can utilize unlabeled samples for unsupervised meta learning, significantly reducing the dependence on a large number of labeled samples while effectively improving the accuracy of HSIs few-shot classification. Firstly, based on the idea of self-supervised learning, multiple different augmentation features of the same sample are generated through principal component analysis and data augmentation methods, to form a large number of different tasks and perform meta-training on the designed model. Then, a few labeled samples randomly selected from target HSIs are used to fine-tune the model, and all labeled samples are used for classification test to evaluate the few-shot classification performance of the model. Moreover, the voting strategy is introduced in the fine-tuning and classification phase, to further improve classification accuracy. The proposed method can perform sufficient unsupervised meta-training under the condition of zero labeled sample, breaking through the bottleneck and limitation of the supervised meta learning methods requiring a large number of source labeled samples. Experiments on three public datasets show that the proposed method can obtain better classification results than existing advanced methods in the few-shot scenario.

Key words: hyperspectral images classification, few-shot classification, unsupervised meta learning, deep learning, voting strategy

中图分类号: