Local-Descriptors-Based Rectification Network for Few-Shot Remote Sensing Scene Classification

Abstract

Few-shot remote sensing scene classification has become a study that has attracted widespread attention and aims to identify new scene classes through one or a few labeled scene images. Nevertheless, due to the existence of unrelated complex background in scene images, local descriptors (LDs) that offer a more efficient representation than image-level features, will carry semantic information unrelated to the real semantics of the scene images. Concurrently, these irrelevant background LDs are also causing a large distribution bias in support and query sets, which leads to the problem of inaccurate feature representation of scene images. To address the aforementioned problems, in this article, we introduce an LD-based rectification network called LDRNet. Within this network, we first design an LD semantic rectification module. It performs semantic rectification on LDs that are unrelated to scene image semantics by obtaining a descriptor-level global-aware semantic representation. Second, we introduce a cross-set bias rectification module. It rectifies the query set by obtaining the offset between two sets (query and support) from a more detailed LD perspective. This operation can shorten the distance among the two sets (query and support), thereby obtaining a more accurate representation of scene image features. Furthermore, we employ an LD-based contrastive loss function to guarantee that the rectified LD semantics are consistent with the corresponding scene image. The comparative experimental result indicates that our LDRNet achieves state-of-the-art performance on three commonly used public datasets.