A Deep-Learning Framework for Land-Sliding Classification from Remote Sensing Image

Despite deep neural networks (DNNs) have been widely applied in remote sensing (RS) scene classification and achieved satisfying performance, the vulnerability of DNNs towards adversarial examples significantly degrades their performance. Moreover, the relatively limited labeled samples of RS scene classification make DNNs more likely to overfit, leading to weak generalizability and noise sensitivity. This may result in DNNs being more vulnerable to adversarial examples. Consequently, the defense of adversarial examples is of crucial importance to improve both the generalizability and robustness of DNNs in the RS scene classification task. However, few studies have been conducted on defense for RS scene classification, especially ignoring the intrinsic characteristics of RS images. In this paper, an effective defense framework for RS scene classification, named reconstruction-assisted and distance-optimized adversarial training (RDAT), is proposed to defend adversarial examples. In order to solve the problems caused by high interclass similarity, a distance-optimized (DO) strategy is designed for adversarial training to strengthen the learning of underfitting content, increase the interclass distance, and improve the robustness of the networks. Furthermore, in order to generate high quality samples for adversarial training, a reconstruction-assisted (RA) block is proposed to eliminate adversarial perturbations in adversarial examples. Specifically, in this block, by swin transformer (SwinT) block and multi-scale convolution (MSC) block, SwinT-MSC-UNet (SMUNet) is constructed to fully extract global and multi-scale local features to adapt to the characteristics of RS images with large variance of ground object scales. Extensive experiments on the benchmark datasets, i.e., UC Merced (UCM) and Aerial Image Dataset (AID), have demonstrate that the proposed RDAT can effectively resist multiple adversarial attacks and yield superior results than other defense methods for RS scene classification.

In recent years, enormous research has been made to improve the classification performance of single-modal remote sensing (RS) data. However, with the ever-growing availability of RS data acquired from satellite or airborne platforms, simultaneous processing and analysis of multimodal RS data pose a new challenge to researchers in the RS community. To this end, we propose a deep-learning-based new framework for multimodal RS data classification, where convolutional neural networks (CNNs) are taken as a backbone with an advanced cross-channel reconstruction module, called CCR-Net. As the name suggests, CCR-Net learns more compact fusion representations of different RS data sources by the means of the reconstruction strategy across modalities that can mutually exchange information in a more effective way. Extensive experiments conducted on two multimodal RS datasets, including hyperspectral (HS) and light detection and ranging (LiDAR) data, i.e., the Houston2013 dataset, and HS and synthetic aperture radar (SAR) data, i.e., the Berlin dataset, demonstrate the effectiveness and superiority of the proposed CCR-Net in comparison with several state-of-the-art multimodal RS data classification methods. The codes will be openly and freely available at <uri xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">https://github.com/danfenghong/IEEE_TGRS_CCR-Net</uri> for the sake of reproducibility.

Limited resolution is one of the most important factors hindering the application of remote sensing images (RSIs). Single-image super resolution (SISR) is a technique to improve the spatial resolution of digital images and has attracted the attention of many researchers. In recent years, with the advancement of deep learning (DL) frameworks, many DL-based SISR models have been proposed and achieved state-of-the-art performance; however, most SISR models for RSIs use the bicubic downsampler to construct low-resolution (LR) and high-resolution (HR) training pairs. Considering that the quality of the actual RSIs depends on a variety of factors, such as illumination, atmosphere, imaging sensor responses, and signal processing, training on “ideal” datasets results in a dramatic drop in model performance on real RSIs. To address this issue, we propose to build a more realistic training dataset by modeling the degradation with blur kernels and imaging noises. We also design a novel residual balanced attention network (RBAN) as a generator to estimate super-resolution results from the LR inputs. To encourage RBAN to generate more realistic textures, we apply a UNet-shape discriminator for adversarial training. Both referenced evaluations on synthetic data and non-referenced evaluations on actual images were carried out. Experimental results validate the effectiveness of the proposed framework, and our model exhibits state-of-the-art performance in quantitative evaluation and visual quality. We believe that the proposed framework can facilitate super-resolution techniques from research to practical applications in RSIs processing.

The evolution of Sixth Generation (6 G) demands advanced sensing frameworks that integrate Remote Sensing (RS), communication, and navigation technologies while addressing challenges related to bandwidth constraints, privacy preservation, and adaptability in dynamic environments. Traditional centralized AI-based RS systems often suffer from inefficiencies, privacy vulnerabilities, and limited generalization. To overcome these limitations, we propose a 6G-compatible framework that unifies multi-task deep learning, Federated Learning (FL), and Explainable AI (XAI) for environmental sensing using the SAT-6 aerial dataset. Our approach begins with a quality-assured preprocessing module, which leverages the Structural Similarity Index (SSIM) and Peak Signal-to-Noise Ratio (PSNR) to enhance image fidelity under noisy conditions. A lightweight convolutional neural network (CNN) is then trained for aerial scene classification, achieving 97% test accuracy while maintaining computational efficiency for edge deployment. To ensure privacy-aware model optimization, we implement FL, allowing decentralized clients to collaboratively train the model without sharing raw data. Our federated setup achieves 97.0% global accuracy within five communication rounds, demonstrating rapid convergence and minimal privacy budget consumption. Additionally, we integrate SHAP-based explainability to interpret model decisions, providing visual explanations that enhance trust and accountability in AI-assisted sensing. This end-to-end framework aligns with 6 G design principles by enabling intelligent, decentralized, and interpretable sensing optimized for real-time deployment in edge-aware and bandwidth-constrained RS and Integrated Sensing and Communication (ISAC) environments.

The classification of ship targets in remote sensing images holds significant application value in fields such as marine monitoring and national defence. Although existing research has yielded considerable achievements in ship classification, current methods struggle to distinguish highly similar ship categories for fine-grained classification tasks due to a lack of targeted design. Specifically, they exhibit the following shortcomings: limited ability to extract locally discriminative features; inadequate fusion of features at high and low levels of representation granularity; and sensitivity of model performance to background noise. To address this issue, this paper proposes a fine-grained classification framework for ship targets in remote sensing images based on Multi-Granularity Attention Network (MA-Net), specifically designed to tackle the aforementioned three major challenges encountered in fine-grained classification tasks for ship targets in remote sensing. This framework first performs multi-level feature extraction through a backbone network, subsequently introducing an Adaptive Local Feature Attention (ALFA) module. This module employs dynamic overlapping region segmentation techniques to assist the network in learning spatial structural combinations, thereby optimising the representation of local features. Secondly, a Dynamic Multi-Granularity Feature Fusion (DMGFF) module is designed to dynamically fuse feature maps of varying representational granularities and select key attribute features. Finally, a Feature-Based Data Augmentation (FBDA) method is developed to effectively highlight target detail features, thereby enhancing feature expression capabilities. On the public FGSC-23 and FGSCR-42 datasets, MA-Net attains top-performing accuracies of 93.12% and 98.40%, surpassing the previous best methods and establishing a new state of the art for fine-grained classification of ship targets in remote sensing images.

Wheat lodging has a significant impact on yields and quality, necessitating the accurate acquisition of lodging information for effective disaster assessment and damage evaluation. This study presents a novel approach for wheat lodging detection in large and heterogeneous fields using UAV remote sensing images. A comprehensive dataset spanning an area of 2.3117 km2 was meticulously collected and labeled, constituting a valuable resource for this study. Through a comprehensive comparison of algorithmic models, remote sensing data types, and model frameworks, this study demonstrates that the Deeplabv3+ model outperforms various other models, including U-net, Bisenetv2, FastSCN, RTFormer, Bisenetv2, and HRNet, achieving a noteworthy F1 score of 90.22% for detecting wheat lodging. Intriguingly, by leveraging RGB image data alone, the current model achieves high-accuracy rates in wheat lodging detection compared to models trained with multispectral datasets at the same resolution. Moreover, we introduce an innovative multi-branch binary classification framework that surpasses the traditional single-branch multi-classification framework. The proposed framework yielded an outstanding F1 score of 90.30% for detecting wheat lodging and an accuracy of 86.94% for area extraction of wheat lodging, surpassing the single-branch multi-classification framework by an improvement of 7.22%. Significantly, the present comprehensive experimental results showcase the capacity of UAVs and deep learning to detect wheat lodging in expansive areas, demonstrating high efficiency and cost-effectiveness under heterogeneous field conditions. This study offers valuable insights for leveraging UAV remote sensing technology to identify post-disaster damage areas and assess the extent of the damage.

Recently, Time Series Remote Sensing Images (TSRSIs) have been proven to be a significant resource for land use/land cover (LULC) mapping. Deep learning methods perform well in managing and processing temporal dependencies and have shown remarkable advancements within this domain. Although deep learning methods have exhibited outstanding performance in classifying TSRSIs, they rely on enough labeled time series samples for effective training. Labeling data with a wide geographical range and a long time span is highly time-consuming and labor-intensive. Active learning (AL) is a promising method of selecting the most informative data for labeling to save human labeling efforts. It has been widely applied in the remote sensing community, except for the classification of TSRSIs. The main challenge of AL in TSRSI classification is dealing with the internal temporal dependencies within TSRSIs and evaluating the informativeness of unlabeled time series data. In this paper, we propose a data-driven active deep learning framework for TSRSI classification to address the problem of limited labeled time series samples. First, a temporal classifier for TSRSI classification tasks is designed. Next, we propose an effective active learning method to select informative time series samples for labeling, which considers representativeness and uncertainty. For representativeness, we use the K-shape method to cluster time series data. For uncertainty, we construct an auxiliary deep network to evaluate the uncertainty of unlabeled data. The features with rich temporal information in the classifier's middle-hidden layers will be fed into the auxiliary deep network. Then, we define a new loss function with the aim of improving the deep model's performance. Finally, the proposed method in this paper was verified on two TSRSI datasets. The results demonstrate a significant advantage of our method over other approaches to TSRSI. On the MUDS dataset, when the initial number of samples was 100 after our method selected and labeled 2000 samples, an accuracy improvement of 4.92% was achieved. On the DynamicEarthNet dataset, when the initial number of samples was 1000 after our method selected and labeled 2000 samples, an accuracy improvement of 7.81% was attained. On the PASTIS dataset, when the initial number of samples was 1000 after our method selected and labeled 2000 samples, an accuracy improvement of 4.89% was achieved. Our code is available in Data Availability Statement.

Cloud is always the weak and even uninformative area inevitably existing in the remote sensing images, and greatly limits the development of remote sensing application. Accurate and automatic detection and segmentation of clouds in satellite scenes is a key problem for the application of remote sensing images. In this paper, we propose a fully convolutional network (FCN) based on deep learning framework for cloud segmentation from remote sensing images. First, the obtained remote sensing image is pre-processed by using band merging and size cropping. Second the image is input into the FCN to achieve the cloud segmentation. To demonstrate the effectiveness of the method, we conduct experiments on Landsat-8 satellite dataset. The overall accuracy of our proposed method for cloud segmentation is higher than 88%. Experimental results demonstrate that both thin and thick cloud can be well segmented with higher accuracy and robustness by using our method.

Multilabel remote sensing (RS) image annotation is a challenging and time-consuming task that requires a considerable amount of expert knowledge. Most existing RS image annotation methods are based on handcrafted features and require multistage processes that are not sufficiently efficient and effective. An RS image can be assigned with a single label at the scene level to depict the overall understanding of the scene and with multiple labels at the object level to represent the major components. The multiple labels can be used as supervised information for annotation, whereas the single label can be used as additional information to exploit the scene-level similarity relationships. By exploiting the dual-level semantic concepts, we propose an end-to-end deep learning framework for object-level multilabel annotation of RS images. The proposed framework consists of a shared convolutional neural network for discriminative feature learning, a classification branch for multilabel annotation and an embedding branch for preserving the scene-level similarity relationships. In the classification branch, an attention mechanism is introduced to generate attention-aware features, and skip-layer connections are incorporated to combine information from multiple layers. The philosophy of the embedding branch is that images with the same scene-level semantic concepts should have similar visual representations. The proposed method adopts the binary cross-entropy loss for classification and the triplet loss for image embedding learning. The evaluations on three multilabel RS image data sets demonstrate the effectiveness and superiority of the proposed method in comparison with the state-of-the-art methods.

This study proposes PlantViT, a Vision Transformer (ViT)-based framework for high-precision vegetation classification by integrating hyperspectral imaging (HSI) and Light Detection and Ranging (LiDAR) data. The dual-branch architecture optimizes feature fusion across spectral and spatial dimensions, where the LiDAR branch extracts elevation and structural features while minimizing information loss and the HSI branch applies involution-based feature extraction to enhance spectral discrimination. By leveraging involution-based feature extraction and a Lightweight ViT (LightViT), the proposed method demonstrates superior classification performance. Experimental results on the Houston 2013 and Trento datasets show that PlantViT achieves an overall accuracy of 99.0% and 97.4%, respectively, with strong agreement indicated by Kappa coefficients of 98.7% and 97.2%. These results highlight PlantViT’s robust capability in classifying heterogeneous vegetation, outperforming conventional CNN-based and other ViT-based models. This study advances Unmanned Aerial Vehicle (UAV)-based remote sensing (RS) for environmental monitoring by providing a scalable and efficient solution for wetland and forest ecosystem assessment.

As wind energy is a kind of clean energy, wind power generation industry has developed rapidly in recent years. Therefore, it is of great significance for the power industry and relevant regulatory authorities to obtain the number and spatial distribution of wind turbine quickly based on remote sensing images. Despite of the rapid development of remote sensing information extraction, the traditional algorithms based on remote sensing classification and recognition cannot meet the requirements of wind turbine extraction on a large-scale area due to the insufficient feature expression of remote sensing images. In this paper, an iterative training method based on the Faster R-CNN (the target extract framework of deep learning) which uses the positive and negative sample set obtained from the first target detection result of ResNet classification for training again has been proposed, which has got the 97.5 percent accuracy for wind turbine target extraction on the 2m spatial resolution map of domestic satellite remote sensing image on the whole china.

Detecting and monitoring landslides are hot topics in remote sensing community, particularly with the development of remote sensing technologies and the significant progress of computer vision. To the best of our knowledge, no study focused on deep learning-based methods for landslide detection on hyperspectral images. We proposes a deep learning framework with constraints to detect landslides on hyperspectral image. The framework consists of two steps. First, a deep belief network is employed to extract the spectral–spatial features of a landslide. Second, we insert the high-level features and constraints into a logistic regression classifier for verifying the landslide. Experimental results demonstrated that the framework can achieve higher overall accuracy when compared to traditional hyperspectral image classification methods. The precision of the landslide detection on the whole image, obtained by the proposed method, can reach 97.91%, whereas the precision of the linear support vector machine, spectral information divergence, and spectral angle match are 94.36%, 84.50%, and 86.44%, respectively. Also, this article reveals that the high-level feature extraction system has a significant potential for landslide detection, especially in multi-source remote sensing.

Change detection with heterogeneous remote sensing images (Hete-CD) plays a significant role in practical applications, particularly in cases where homogenous remote sensing images are unavailable. However, directly comparing bitemporal heterogeneous remote sensing images (HRSIs) to measure the change magnitude is unfeasible. Numerous deep learning methods require substantial samples to train the module adequately. Moreover, the process of labeling a large number of samples for land cover change detection using HRSIs is time-consuming and labor-intensive. Consequently, deep learning networks face challenges in achieving satisfactory performance in Hete-CD due to the limited number of training samples. This study proposes a novel deep-learning framework for Hete-CD to achieve satisfactory performance even with a limited number of initial samples. We developed a multiscale network with a selected kernel-attention module. This design allows us to effectively capture different change targets characterized by diverse sizes and shapes. In addition, a simple yet effective non-parameter sample-enhanced algorithm that utilizes the Pearson correlation coefficient is proposed to explore the potential samples surrounding every initial sample. The proposed network and sample-enhanced algorithm are integrated into an iterative framework to improve change detection performance with a limited number of small samples. The experimental results were achieved based on four pairs of real HRSIs, which were acquired with Landsat-5, Radarsat-2, and Sentinel-2 satellites with optical and SAR sensors. Results indicated that the proposed framework could achieve competitive accuracy with a small number of samples compared with some state-of-the-art methods, including three traditional methods and nine state-of-the-art deep learning methods. For example, the improvement rates are approximately 3.38% and 1.99% compared with the selected traditional methods and deep learning methods, respectively.

Multi-label classification of remote sensing (RS) images under the deep learning framework has attracted increasing attentions. However, manual annotation of RS images is time-consuming and easily results in missing labels due to the subjectivity of human perception. The deep learning models could suffer from performance drop because of the incomplete labels of the training set. To address the issue, a deep learning model with label enhancement is proposed for incomplete multi-label classification of RS images. The model consists of a semantic feature extraction (SFE) module, a label correlation learning (LCL) module and a label enhancement (LE) module. Firstly, the semantic features of RS images are extracted by an improved ResNet-101 in the SFE module. Secondly, the global label correlations are learned by a graph convolutional network (GCN) in the LCL module. Subsequently, the labels predicted according to the semantic features are enhanced by the label correlations and local semantics in the LE module. Finally, a weighted combination of the label probability generated by the semantic features and the enhanced one through the LE module is applied in the loss function for classification. The multi-label classification experiments on three public RS image datasets prove the effectiveness of our model.

Remote sensing imagery inherently contains geographic coordinate information, and the spatial distribution of land-cover types often exhibits pronounced regional heterogeneity. However, most deep learning approaches in remote sensing still follow computer vision paradigms, treating images as ordinary pixel matrices while ignoring their intrinsic geographic attributes, which limits both representation capacity and spatial generalization. To address this issue, we propose a spatially enhanced deep learning framework that explicitly incorporates geographic reference information. Specifically, the geographic coordinates of image centers are introduced as additional input channels into convolutional neural networks through a simple embedding mechanism, enabling the model to adaptively capture regional variations while maintaining a unified and backbone-agnostic architecture. Based on the EuroSAT dataset, we construct a new benchmark, termed Geo-EuroSAT, by embedding coordinate information and conduct systematic comparative experiments under multiple training–validation split strategies. Both ResNet-34 and EfficientNet-B3 backbones are evaluated to verify the general applicability of the proposed method. Experimental results demonstrate that the coordinate-aware models consistently outperform conventional baselines across all settings (p < 0.005), with overall accuracy improvements ranging from 0.4% to 1.8%. The gains are especially pronounced under limited training samples and for vegetation-related classes such as <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Pasture</i> and <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Herbaceous Vegetation</i>. These findings indicate that explicitly integrating geographic reference into deep networks via a lightweight, backbone-agnostic embedding provides a principled and generalizable way to enhance large-scale land-cover classification and geographic process modeling.