Easy-to-Hard Domain Adaptation With Human Interaction for Hyperspectral Image Classification

In real-world hyperspectral image classification (HSIC), the limited annotated examples usually lead to an insufficiently trained classifier, which further generates low classification accuracy. To overcome this challenge, domain adaptation (DA) methods have been developed to transfer learnable knowledge from the external HSIs with sufficient labeled examples (i.e., the source domain) to the interested HSIs with scarce labeled examples (i.e., the target domain). Conventional DA approaches often pseudo-label the examples with high classification confidence and then incorporate them into the training process. However, due to the significant domain gap, relying solely on confident examples may not be adequate to achieve satisfactory performance. Therefore, this article proposes an interactive easy-to-hard DA method (IEH-DA) to arrange the adaptation process so that the "easy" examples are adapted ahead of the "hard" ones. In an early stage, the easy examples with high pseudo-labeling confidence are selected for the adversarial learning-based DA. In a later stage, the "hard" examples with high informativity are further selected, and they are interactively labeled by human experts to provide accurate supervision information for adaptation. As a result, the examples in the target domain are used in an easy-to-hard way, which forms a curriculum sequence for orderly model training. Extensive experiments conducted on typical public datasets demonstrate that IEH-DA outperforms other state-of-the-art DA methods for HSIC.