Clouds and snow in remote sensing imageries cover underlying surface information, reducing image availability. Moreover, they interact with each other, decreasing the cloud and snow detection accuracy. In this study, we propose a convolutional neural network for cloud and snow detection, named the cloud and snow detection network (CSD-Net). It incorporates the multi-scale feature fusion module (MFF) and the controllably deep supervision and feature fusion structure (CDSFF). MFF can capture and aggregate features at various scales, ensuring that the extracted high-level semantic features of clouds and snow are more distinctive. CDSFF can provide a deeply supervised mechanism with hinge loss and combine information from adjacent layers to gain more representative features. It ensures the gradient flow is more oriented and error-less, while retaining more effective information. Additionally, a high-resolution cloud and snow dataset based on WorldView2 (CSWV) was created and released. This dataset meets the training requirements of deep learning methods for clouds and snow in high-resolution remote sensing images. Based on the datasets with varied resolutions, CSD-Net is compared to eight state-of-the-art deep learning methods. The experiment results indicate that CSD-Net has an excellent detection accuracy and efficiency. Specifically, the mean intersection over the union (MIoU) of CSD-Net is the highest in the corresponding experiment. Furthermore, the number of parameters in our proposed network is just 7.61 million, which is the lowest of the tested methods. It only has 88.06 GFLOPs of floating point operations, which is less than the U-Net, DeepLabV3+, PSPNet, SegNet-Modified, MSCFF, and GeoInfoNet. Meanwhile, CSWV has a higher annotation quality since the same method can obtain a greater accuracy on it.

Deep learning approaches have been widely used in building automatic extraction tasks and have made great progress in recent years. However, the missing detection and wrong detection causing by spectrum confusion is still a great challenge. The existing fully convolutional networks (FCNs) cannot effectively distinguish whether the feature differences are from one building or the building and its adjacent non-building objects. In order to overcome the limitations, a building multi-feature fusion refined network (BMFR-Net) was presented in this paper to extract buildings accurately and completely. BMFR-Net is based on an encoding and decoding structure, mainly consisting of two parts: the continuous atrous convolution pyramid (CACP) module and the multiscale output fusion constraint (MOFC) structure. The CACP module is positioned at the end of the contracting path and it effectively minimizes the loss of effective information in multiscale feature extraction and fusion by using parallel continuous small-scale atrous convolution. To improve the ability to aggregate semantic information from the context, the MOFC structure performs predictive output at each stage of the expanding path and integrates the results into the network. Furthermore, the multilevel joint weighted loss function effectively updates parameters well away from the output layer, enhancing the learning capacity of the network for low-level abstract features. The experimental results demonstrate that the proposed BMFR-Net outperforms the other five state-of-the-art approaches in both visual interpretation and quantitative evaluation.

Current automatic shadow compensation methods often suffer because their contrast improvement processes are not self-adaptive and, consequently, the results they produce do not adequately represent the real objects. The study presented in this paper designed a new automatic shadow compensation framework based on improvements to the Wallis principle, which included an intensity coefficient and a stretching coefficient to enhance contrast and brightness more efficiently. An automatic parameter calculation strategy also is a part of this framework, which is based on searching for and matching similar feature points around shadow boundaries. Finally, a final compensation combination strategy combines the regional compensation with the local window compensation of the pixels in each shadow to improve the shaded information in a balanced way. All these strategies in our method work together to provide a better measurement for customizing suitable compensation depending on the condition of each region and pixel. The intensity component I also is automatically strengthened through the customized compensation model. Color correction is executed in a way that avoids the color bias caused by over-compensated component values, thereby better reflecting shaded information. Images with clouds shadows and ground objects shadows were utilized to test our method and six other state-of-the-art methods. The comparison results indicate that our method compensated for shaded information more effectively, accurately, and evenly than the other methods for customizing suitable models for each shadow and pixel with reasonable time-cost. Its brightness, contrast, and object color in shaded areas were approximately equalized with non-shaded regions to present a shadow-free image.

Deep learning techniques have achieved significant success in landslide mapping (LM) from remote sensing images. However, the need for large quantities of high-quality labels presents a substantial barrier to the practical application of deep learning models. Semi-supervised learning (SSL) offers a solution by leveraging the knowledge contained in unlabeled remote sensing landslide images (RSLIs) to improve model performance. The main challenges in applying SSL to LM include the limitations and appropriateness of consistency perturbation perspectives, as well as the potential disruption of both shallow and deep landslide features and micro-topographic information. To address these challenges, we propose a reliable multi-perspective consistency SSL framework called Semi-LM. This framework employs multi-perspective consistency learning (MCL) as the learning pathway and introduces the dual-branch anchor CutMix (DACM) method and feature perturbation assign (FPA) strategy for LM. MCL achieves unified multi-perspective perturbations by enforcing the alignment of multi-perspective perturbed LM results with generated weak perturbation pseudolabels, expanding the perturbation space and enabling model to learn more discriminative features from unlabeled RSLIs. DACM generates anchor masks by determining NO-Cut zones with landslide anchors, mitigating the risk of erroneous knowledge accumulation and improving the reliability of model optimization. Additionally, by measuring the local variability of the model within a high-dimensional feature space, FPA identifies the specific stage where low-density regions of separated classes are located. Extensive comparative and ablation experiments on five diverse landslide datasets demonstrated that Semi-LM significantly reduces the reliance on labels in LM, identifies more landslides, and provides more stable results.

The scarcity of high-quality labeled data poses a challenge to the application of deep learning (DL) in landslide identification from remote sensing (RS) images. Semi-supervised learning (SSL) has emerged as a promising approach to address the issue of low accuracy caused by the limited availability of high-quality labels. Nevertheless, the application of SSL approaches developed for natural images to landslide identification encounters several challenges. This study focuses on two specific challenges: inadequate information extraction from limited unlabeled RS landslide images and the generation of low-quality pseudo-labels. To tackle these challenges, we propose a novel and generic DL framework called hybrid perturbation mean match (HPM-Match). The framework combines dual-branch input perturbation (DIP) and independent triple-stream perturbation (ITP) techniques to enhance model accuracy with limited labels. The DIP generation approach is designed to maximize the utilization of manually pre-defined perturbation spaces while minimizing the introduction of erroneous information during the weak-to-strong consistency learning (WSCL) process. Moreover, the ITP structure unifies input, feature, and model perturbations, thereby broadening the perturbation space and enabling knowledge extraction from unlabeled landslide images across various perspectives. Experimental results demonstrate that HPM-Match has substantial improvements in IoU, with maximum increases of 26.68%, 7.05%, and 12.96% over supervised learning across three datasets with the same label ratio and reduces the number of labels by up to about 70%. Furthermore, HPM-Match strikes a better balance between precision and recall, identifying more landslides than other state-of-the-art (SOTA) SSL approaches.