Regarding landslide inventory mapping (LIM) as a task similar to change detection, current methods for LIM using bi-temporal optical remote sensing images are generally derived from change detection methods. In practice, not all changed regions belong to landslides, e.g., new roads, canals, and vegetation. Therefore, an ideal strategy is supposed to present two steps: discriminating changed and unchanged regions, and detecting landslides apart from other changed regions. Owing to the complexity and uncertainty of landslides, it is difficult to simultaneously separate landslides with unchanged and other changed regions by a single model. Addressing this problem, in this letter, we apply a generative adversarial network (GAN) in a Siamese neural network, and then propose a GAN-based Siamese framework (GSF) for LIM. The GSF comprises two cascaded modules, namely, domain adaptation and landslide detection. The former module aims to make a cross-domain mapping between prelandslide and postlandslide images with adversarial learning, then translate paired images into the same domain to suppress the domain discrepancies of bi-temporal remote sensing images. Meanwhile, the latter module aims to perform pixel-level landslide detection with a Siamese model. By training this cascaded framework, our method learns to produce landslide inventory maps without any preprocessing or postprocessing. Extensive experiments and comparison with other state-of-the-art methods verify the efficiency and superiority of our method.

As a fundamental and profound task in remote sensing, change detection from very-high-resolution (VHR) images plays a vital role in a wide range of applications and attracts considerable attention. Current methods generally focus on the research of simultaneously modeling and discriminating the changed and unchanged features. In practice, for bi-temporal VHR optical remote sensing images, the temporal spectral variability tends to exist in all bands throughout the entire paired images, making it difficult to distinguish none-changes and changes with a single model. In this paper, motivated by this observation, we propose a novel hybrid end-to-end framework named dual learning-based Siamese framework (DLSF) for change detection. The framework comprises two parallel streams which are dual learning-based domain transfer and Siamese-based change decision. The former stream is aimed at reducing the domain differences of two paired images and retaining the intrinsic information by translating them into each other’s domain. While the latter stream is aimed at learning a decision strategy to decide the changes in two domains, respectively. By training our proposed framework with certain change map references, this method learns a cross-domain translation in order to suppress the differences of unchanged regions and highlight the differences of changed regions in two domains, respectively, then focus on the detection of changed regions. To the best of our knowledge, the idea of incorporating dual learning framework and Siamese network for change detection is novel. The experimental results on two datasets and the comparison with other state-of-the-art methods verify the efficiency and superiority of our proposed DLSF.

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.

With the introduction of artificial intelligence (AI), current deep learning-driven change detection methods generally regard changes as a type of specific land cover information and try to detect them simply by using existing semantic labeling models. In practice, the changes that occur on different land cover categories may appear completely different for remote sensing images, making it difficult to detect multiple categories of changes with an end-to-end model. In addition, training such networks requires a large amount of prelabeled references, which are labor-intensive and time-consuming. Motivated by this observation, we integrate dual learning algorithm and disentangled representation theory to develop a novel approach, named content-invariant dual learning (CiDL), for either supervised or unsupervised change detection in remote sensing images. In our framework, two opposite Y-shaped networks, each of which consists of two encoders and one decoder, are introduced to translate bitemporal images from their original domains to each others, where their intrinsic content features are retained, while their style features are consistent. By training our hybrid framework, even without references, this method learns a category-wise cross-domain translation to suppress the discrepancies in paired unchanged regions, and meanwhile highlight those in paired changed regions. The experimental results on two typical change detection data sets and the comparison with other state-of-the-art deep learning-driven methods verify the effectiveness and competitiveness of our proposed CiDL.

The development of artificial intelligence technology has prompted an immense amount of researches on improving the performance of change detection approaches. Existing deep learning-driven methods generally regard changes as a specific type of land cover, and try to identify them relying on the powerful expression capabilities of neural networks. However, in practice, different types of land cover changes are generally influenced by environmental factors at different degrees. Furthermore, seasonal variation-induced spectral differences seriously interfere with those of real changes in different land cover types. All these problems pose great challenges for season-varying change detection because the real and seasonal variation-induced changes are technically difficult to separate by a single end-to-end model. In this paper, by embedding a convolutional long short-term memory (ConvLSTM) network into a conditional generative adversarial network (cGAN), we develop a novel method, named progressive domain adaptation (PDA), for change detection using season-varying remote sensing images. In our idea, two cascaded modules, progressive translation and group discrimination, are introduced to progressively translate pre-event images from their own domain to the post-event one, where their seasonal features are consistent and their intrinsic land cover distribution features are retained. By training this hybrid multi-model framework with certain reference change maps, the seasonal variation-induced changes between paired images are effectively suppressed, and meanwhile the natural and human activity-caused changes are greatly emphasized. Extensive experiments on two types of season-varying change detection datasets and a comparison with other state-of-the-art methods verify the effectiveness and competitiveness of our proposed PDA.

As the fastest growing trend in big data analysis, deep learning technology has proven to be both an unprecedented breakthrough and a powerful tool in many fields, particularly for image segmentation tasks. Nevertheless, most achievements depend on high-quality pre-labeled training samples, which are labor-intensive and time-consuming. Furthermore, different from conventional natural images, coastal remote sensing ones generally carry far more complicated and considerable land cover information, making it difficult to produce pre-labeled references for supervised image segmentation. In our research, motivated by this observation, we take an in-depth investigation on the utilization of neural networks for unsupervised learning and propose a novel method, namely conditional co-training (CCT), specifically for truly unsupervised remote sensing image segmentation in coastal areas. In our idea, a multi-model framework consisting of two parallel data streams, which are superpixel-based over-segmentation and pixel-level semantic segmentation, is proposed to simultaneously perform the pixel-level classification. The former processes the input image into multiple over-segments, providing self-constrained guidance for model training. Meanwhile, with this guidance, the latter continuously processes the input image into multi-channel response maps until the model converges. Incentivized by multiple conditional constraints, our framework learns to extract high-level semantic knowledge and produce full-resolution segmentation maps without pre-labeled ground truths. Compared to the black-box solutions in conventional supervised learning manners, this method is of stronger explainability and transparency for its specific architecture and mechanism. The experimental results on two representative real-world coastal remote sensing datasets of image segmentation and the comparison with other state-of-the-art truly unsupervised methods validate the plausible performance and excellent efficiency of our proposed CCT.