Foreground-Background Contrastive Learning for Few-Shot Remote Sensing Image Scene Classification

Jie Geng, Bohan Xue, Wen Jiang

Research output: Contribution to journalArticlepeer-review

13 Scopus citations

Abstract

Few-shot learning (FSL) aims to train a model with limited samples for identifying novel category samples. As for remote sensing (RS) images, complex backgrounds may lead to large intraclass differences, and the number of labeled samples is quite smaller than that of large datasets, which both influence the classification performance. To solve these issues, a foreground-background contrastive learning (FBCL) is proposed for few-shot RS image scene classification. Specifically, a foreground-background separation (FBS) module is proposed to separate features between objects and background with supervised contrastive learning (SCL), which aims to improve the ability to distinguish foreground and background regions of RS images. Moreover, a channel weight allocator is proposed to balance features of different dimensions, which can take full advantage of RS image information. Experiments on three RS datasets prove that the proposed few-shot method is able to produce superior classification results than other related approaches.

Original languageEnglish
Article number5614112
JournalIEEE Transactions on Geoscience and Remote Sensing
Volume61
DOIs
StatePublished - 2023

Keywords

  • Contrastive learning
  • few-shot learning (FSL)
  • remote sensing (RS) image
  • scene classification

Fingerprint

Dive into the research topics of 'Foreground-Background Contrastive Learning for Few-Shot Remote Sensing Image Scene Classification'. Together they form a unique fingerprint.

Cite this