State-of-the-art and gaps for deep learning on limited training data in remote sensing
This is an incremental review article for researchers in remote sensing facing data scarcity issues.
The paper reviews three deep learning approaches—transfer learning, unsupervised learning, and generative adversarial networks—to address the challenge of limited training data in remote sensing, highlighting current gaps without presenting new results.
Deep learning usually requires big data, with respect to both volume and variety. However, most remote sensing applications only have limited training data, of which a small subset is labeled. Herein, we review three state-of-the-art approaches in deep learning to combat this challenge. The first topic is transfer learning, in which some aspects of one domain, e.g., features, are transferred to another domain. The next is unsupervised learning, e.g., autoencoders, which operate on unlabeled data. The last is generative adversarial networks, which can generate realistic looking data that can fool the likes of both a deep learning network and human. The aim of this article is to raise awareness of this dilemma, to direct the reader to existing work and to highlight current gaps that need solving.