Domain Knowledge Decomposition for Cross-Domain Few-Shot Scene Classification from Remote Sensing Imagery

Cross-domain few-shot scene classification (CDFSSC) aims to tackle the challenge of classifying target domain data with limited labeled samples under distribution shift and category mismatch in remote sensing (RS) imagery classification. Existing methods primarily focus on extracting domain-common knowledge while overlooking domain-specific knowledge, which is insufficient for effective cross-domain representation learning. In addition, the category mismatch between the source and target domains further hinders the model's ability to adapt and perform effectively on few-shot tasks in the target domain. Hence, in this article, a novel CDFSSC method called domain knowledge decomposition (DKD) framework is proposed to effectively exploit domain-common and domain-specific knowledge from the pseudo-labels of target samples, improve the certainty of cross-domain representation learning, and enhance the model's adaptability to the target domain. First, a cross-domain pseudo-label decomposed learning structure is proposed for DKD to facilitate domain-common knowledge transfer and correct interference with domain-specific knowledge. It decomposes the pseudo-labels at the logit level to separately exploit two types of knowledge, ensuring a more effective and robust cross-domain representation learning. Second, a certainty-enhanced dynamic loss is designed to strengthen the certainty in cross-domain representation learning by minimizing the self-entropy of predictions and the dynamic loss reweighting principle. Third, in the target domain adaptation and few-shot evaluation stage, a target-domain-specific adapter is designed to improve the model's adaptability to target domain few-shot tasks, while addressing category mismatch by classifier fine-tuning. Extensive experimental results on 12 RS cross-domain scenarios exhibit impressive performance of the proposed DKD framework.