Abstract
Since manually labeling aerial images for pixel-level classification is expensive and time-consuming, developing strategies for land cover mapping without reference labels is essential and meaningful. As an efficient solution for this issue, domain adaptation has been widely utilized in numerous semantic labeling-based applications. However, current approaches generally pursue the marginal distribution alignment between the source and target features and ignore the category-level alignment. Therefore, directly applying them to land cover mapping leads to unsatisfactory performance in the target domain. In our research, to address this problem, we embed a geometry-consistent generative adversarial network (GcGAN) into a co-training adversarial learning network (CtALN), and then develop a category-sensitive domain adaptation (CsDA) method for land cover mapping using very-high-resolution (VHR) optical aerial images. The GcGAN aims to eliminate the domain discrepancies between labeled and unlabeled images while retaining their intrinsic land cover information by translating the features of the labeled images from the source domain to the target domain. Meanwhile, the CtALN aims to learn a semantic labeling model in the target domain with the translated features and corresponding reference labels. By training this hybrid framework, our method learns to distill knowledge from the source domain and transfers it to the target domain, while preserving not only global domain consistency, but also category-level consistency between labeled and unlabeled images in the feature space. The experimental results between two airborne benchmark datasets and the comparison with other state-of-the-art methods verify the robustness and superiority of our proposed CsDA.
Highlights
Land cover, a fundamental variable considering both natural and artificial surface structures, plays important roles in various scientific studies, such as climate change, resource investigations, and sustainable development [1,2,3]
As one of the most accessible and used remote sensing types of data, current very-high-resolution (VHR) optical aerial images provide a wealth of detailed information about land cover categories due to their wide coverage and high resolution; they are characterized by redundancy and noise
Land cover mapping is a type of pixel-level image classification task, and current approaches can be divided into two primary classes: with reference and without reference
Summary
A fundamental variable considering both natural and artificial surface structures, plays important roles in various scientific studies, such as climate change, resource investigations, and sustainable development [1,2,3]. Land cover mapping is a type of pixel-level image classification task, and current approaches can be divided into two primary classes: with reference and without reference. The former methods focus on learning semantic labeling models with guidance by certain references, which are manually pre-labeled reference land cover maps. The latter methods focus on that with guidance by prior knowledge obtained using the human visual system
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.