Unsupervised Few-Shot Continual Learning for Remote Sensing Image Scene Classification

TL;DR

This paper introduces UNISA, an unsupervised few-shot continual learning method for remote sensing image classification that does not require labeled data, effectively handling data scarcity and catastrophic forgetting.

Contribution

The paper proposes UNISA, a novel unsupervised continual learning approach for remote sensing, combining prototype scattering, positive sampling, and a flat-wide learning strategy.

Findings

UNISA outperforms existing methods on remote sensing datasets.

The approach effectively addresses data scarcity and catastrophic forgetting.

Source code is publicly available for reproducibility.

Abstract

A continual learning (CL) model is desired for remote sensing image analysis because of varying camera parameters, spectral ranges, resolutions, etc. There exist some recent initiatives to develop CL techniques in this domain but they still depend on massive labelled samples which do not fully fit remote sensing applications because ground truths are often obtained via field-based surveys. This paper addresses this problem with a proposal of unsupervised flat-wide learning approach (UNISA) for unsupervised few-shot continual learning approaches of remote sensing image scene classifications which do not depend on any labelled samples for its model updates. UNISA is developed from the idea of prototype scattering and positive sampling for learning representations while the catastrophic forgetting problem is tackled with the flat-wide learning approach combined with a ball generator to address the data scarcity problem. Our numerical study with remote sensing image scene datasets and a hyperspectral dataset confirms the advantages of our solution. Source codes of UNISA are shared publicly in \url{https://github.com/anwarmaxsum/UNISA} to allow convenient future studies and reproductions of our numerical results.