Domain Adaptive Remote Sensing Scene Recognition via Semantic Relationship Knowledge Transfer

Scene recognition has attracted rising attentions of many researchers in the remote sensing fields, owing to the rapidly advancing of remote sensing devices in recent years. However, images obtained from various sensors dominate diverse sensor-specific characteristics, which will dramatically weaken the model transferability trained on a source data domain to a different target domain on account of the domain shift issues. To mitigate the domain discrepancy, most existing methods attend to align the cross-domain distributions. While the valuable knowledge of semantic relationships between different scenes is generally overlooked, and the underlying correlation across scenes cannot be fully discovered. For the sake of tackling this challenge, we propose an adaptive remote sensing scene recognition network, which can successfully transfer both the discriminative knowledge and cross-scene relationship from source to target. Specifically, in this article, we acquire sensor-invariant representations in an adversarial manner and realize fine-grained conditional distribution alignment contrastively. In such a way, the tremendous domain gap can be mitigated to a large extent, and the discriminative and well-matched representations will be derived favorably. In addition, we explicitly construct classwise relationship distributions belonging to two domains, respectively, and minimize their divergence to conduct semantic relationship knowledge transfer (SRKT), for the purpose of sufficiently unearthing the intrinsic semantic relative structures that can prompt generality of the model in the target domain. Finally, we conduct multiple experiments on representative multidomain remote sensing benchmarks, and the extensive experimental results demonstrate the superiority of our proposed approach.

Automated summary

Scene recognition in remote sensing has gained increasing attention due to advances in remote sensing devices. However, the domain shift problem caused by the diverse sensor-specific characteristics of images obtained from various sensors weakens the transferability of models trained on one data domain to a different target domain. To address this challenge, we propose an adaptive remote sensing scene recognition network that transfers both discriminative knowledge and cross-scene relationship from source to target. Our approach aligns the distributions of different domains, discovers semantic relationships between scenes, and achieves superior performance on remote sensing benchmarks.