Search Results (91)

High-resolution remote sensing (HRRS) imagery enables the extraction of cropland information with high levels of detail, especially when combined with the impressive performance of deep convolutional neural networks (DCNNs) in understanding these images. Comprehending the factors influencing DCNNs’ performance in HRRS cropland extraction is of considerable importance for practical agricultural monitoring applications. This study investigates the impact of classifier selection and different training data characteristics on the HRRS cropland classification outcomes. Specifically, Gaofen-1 composite images with 2 m spatial resolution are employed for HRRS cropland extraction, and two county-wide regions with distinct agricultural landscapes in Shandong Province, China, are selected as the study areas. The performance of two deep learning (DL) algorithms (UNet and DeepLabv3+) and a traditional classification algorithm, Object-Based Image Analysis with Random Forest (OBIA-RF), is compared. Additionally, the effects of different band combinations, crop growth stages, and class mislabeling on the classification accuracy are evaluated. The results demonstrated that the UNet and DeepLabv3+ models outperformed OBIA-RF in both simple and complex agricultural landscapes, and were insensitive to the changes in band combinations, indicating their ability to learn abstract features and contextual semantic information for HRRS cropland extraction. Moreover, compared with the DL models, OBIA-RF was more sensitive to changes in the temporal characteristics. The performance of all three models was unaffected when the mislabeling error ratio remained below 5%. Beyond this threshold, the performance of all models decreased, with UNet and DeepLabv3+ showing similar performance decline trends and OBIA-RF suffering a more drastic reduction. Furthermore, the DL models exhibited relatively low sensitivity to the patch size of sample blocks and data augmentation. These findings can facilitate the design of operational implementations for practical applications. Full article

Hyperspectral sparse unmixing (SU) networks have recently received considerable attention due to their model hyperspectral images (HSIs) with a priori spectral libraries and to capture nonlinear features through deep networks. This method effectively avoids errors associated with endmember extraction, and enhances the unmixing performance via nonlinear modeling. However, two major challenges remain: the use of large spectral libraries with high coherence leads to computational redundancy and performance degradation; moreover, certain feature extraction models, such as Transformer, while exhibiting strong representational capabilities, suffer from high computational complexity. To address these limitations, this paper proposes a hyperspectral unmixing dual-branch network integrating an adaptive hop-aware GCN and neighborhood offset Mamba that is termed DGMNet. Specifically, DGMNet consists of two parallel branches. The first branch employs the adaptive hop-neighborhood-aware GCN (AHNAGC) module to model global spatial features. The second branch utilizes the neighborhood spatial offset Mamba (NSOM) module to capture fine-grained local spatial structures. Subsequently, the designed Mamba-enhanced dual-stream feature fusion (MEDFF) module fuses the global and local spatial features extracted from the two branches and performs spectral feature learning through a spectral attention mechanism. Moreover, DGMNet innovatively incorporates a spectral-library-pruning mechanism into the SU network and designs a new pruning strategy that accounts for the contribution of small-target endmembers, thereby enabling the dynamic selection of valid endmembers and reducing the computational redundancy. Finally, an improved ESS-Loss is proposed, which combines an enhanced total variation (ETV) with an $l_{1/2}$ sparsity constraint to effectively refine the model performance. The experimental results on two synthetic and five real datasets demonstrate the effectiveness and superiority of the proposed method compared with the state-of-the-art methods. Notably, experiments on the Shahu dataset from the Gaofen-5 satellite further demonstrated DGMNet’s robustness and generalization. Full article

In January 2025, a magnitude 6.8 earthquake struck Dingri County, Shigatse, Tibet, causing severe damage. Rapid and precise extraction of damaged buildings is essential for emergency relief and rebuilding efforts. This study proposes an approach integrating YOLO-E (Real-Time Seeing Anything) and the Segment Anything Model 2 (SAM2) to extract damaged buildings with multi-source remote sensing images, including post-earthquake Gaofen-7 imagery (0.80 m), Beijing-3 imagery (0.30 m), and pre-earthquake Google satellite imagery (0.15 m), over the affected region. In this hybrid approach, YOLO-E functions as the preliminary segmentation module for initial segmentation. It leverages its real-time detection and segmentation capability to locate potential damaged building regions and generate coarse segmentation masks rapidly. Subsequently, SAM2 follows as a refinement step, incorporating shapefile information from pre-disaster sources to apply precise, pixel-level segmentation. The dataset used for training contained labeled examples of damaged buildings, and the model optimization was carried out using stochastic gradient descent (SGD), with cross-entropy and mean squared error as the selected loss functions. Upon evaluation, the model reached a precision of 0.840, a recall of 0.855, an F1-score of 0.847, and an IoU of 0.735. It successfully extracted 492 suspected damaged building patches within a radius of 20 km from the earthquake epicenter, clearly showing the distribution characteristics of damaged buildings concentrated in the earthquake fault zone. In summary, this hybrid YOLO-E and SAM2 approach, leveraging multi-source remote sensing imagery, delivers precise and rapid extraction of damaged buildings with a precision of 0.840, recall of 0.855, and IoU of 0.735, effectively supporting targeted earthquake rescue and post-disaster reconstruction efforts in the Dingri County fault zone. Full article

Influenced by a warm and humid climate, the permafrost on the Qinghai–Tibet Plateau is undergoing significant degradation, leading to the occurrence of extensive thermokarst landforms. Among the most typical landforms in permafrost areas is thaw slump. This study, based on three periods of data from keyhole images of 1968–1970, the fractional images of 2006–2009 and the Gaofen (GF) images of 2018–2019, combined with field surveys for validation, investigates the distribution characteristics and spatiotemporal variation trends of thaw slumps in the Hoh Xil area and evaluates the susceptibility to thaw slumping in this area. The results from 1968 to 2019 indicate a threefold increase in the number and a twofold increase in total area of thaw slumps. Approximately 70% of the thaw slumps had areas less than 2 × 10⁴ m². When divided into a grid of 3 km × 3 km, about 1.3% (128 grids) of the Hoh Xil region experienced thaw slumping from 1968 to 1970, while 4.4% (420 grids) showed such occurrences from 2018 to 2019. According to the simulation results obtained using the informativeness method, the area classified as very highly susceptible to thaw slumping covers approximately 26% of the Hoh Xil area, while the highly susceptible area covers about 36%. In the Hoh Xil, 61% of the thaw slump areas had an annual warming rate ranging from 0.18 to 0.25 °C/10a, with 70% of the thaw slump areas experiencing a precipitation increase rate exceeding 12 mm/10a. Future assessments of thaw slump development suggest a possible minimum of 41 and a maximum of 405 thaw slumps occurrences annually in the Hoh Xil region. Under rapidly changing climatic conditions, apart from environmental risks, there also exist substantial potential risks associated with thaw slumping, such as the triggering of large-scale landslides and debris flows. Therefore, it is imperative to conduct simulated assessments of thaw slumping throughout the entire plateau to address regional risks in the future. Full article

The estimation and correction of antenna patterns are essential for ensuring the relative radiometric quality of SAR images. Traditional methods for antenna pattern estimation rely on artificial calibrators or specific stable natural scenes like the Amazon rainforest, which are limited by cost, complexity, and geographic constraints, making them unsuitable for frequent imaging demands. Meanwhile, general natural scenes are imaged frequently using SAR systems, but their true scattering characteristics are unknown, posing a challenge for direct antenna pattern estimation. Therefore, it is considered to use the calibrated SAR to obtain the scattering characteristics of these general scenarios; that is, introducing the concept of cross-calibration. Accordingly, this paper proposes a novel method for estimating the SAR range antenna pattern based on cross-calibration. The method addresses three key challenges: (1) Identifying pseudo-invariant natural scenes suitable as reference targets through spatial uniformity and temporal stability assessments using standard deviation and amplitude correlation analyses; (2) Achieving pixel-level registration of heterogeneous SAR images with an iterative method despite radiometric imbalances; (3) Extracting stable power values by segmenting images and applying differential screening to minimize outlier effects. The proposed method is validated using Gaofen-3 SAR data and shows robust performance in bare soil, grassland, and forest scenarios. Comparing the results of the proposed method with the tropical forest-based calibration method, the maximum shape deviation between the range antenna patterns of the two methods is less than 0.2 dB. Full article

The Directional Polarimetric Camera (DPC) aboard the Chinese GaoFen-5 02 satellite is designed to monitor aerosols and particulate matter (PM). In this study, we retrieved the aerosol optical depth (AOD) over the Jing–Jin–Ji (JJJ) region using multi-angle data from the DPC, employing a combination of dark dense vegetation (DDV) and multi-angle retrieval methods. The added value of our method included novel hybrid methodology and good practical performance. The retrieval process involves three main steps: (1) deriving AOD from DPC data collected at the nadir angle using linear parameters of land surface reflectance between the blue and red bands from the MOD09 surface product; (2) after performing atmospheric correction with the retrieved AOD, calculating the variance of the normalized reflectance at all observation angles; and (3) leveraging the calculated variance to obtain the final AOD values. AOD images over the JJJ region were successfully retrieved from DPC data collected between January and June 2022. To validate the retrieval method, we compared our results with aerosol products from the AErosol RObotic NETwork (AERONET) Beijing-RADI site, as well as aerosol data from MODerate-resolution Imaging Spectroradiometer (MODIS) and the generalized retrieval of atmosphere and surface properties (GRASP)/models over the same site. In terms of validation metrics, the correlation coefficient (R²) and root mean square error (RMSE) indicated that our method achieved high accuracy, with an R² value greater than 0.9 and an RMSE below 0.1, closely aligning with the performance of GRASP. Full article

Volume of change provides a comprehensive and objective reflection of land surface transformation, meeting the emerging demand for feature change monitoring in the era of big data. However, existing land surface monitoring methods often focus on a single dimension, either horizontal or vertical, making it challenging to achieve quantitative volumetric change monitoring. Accurate volumetric change measurements are indispensable in many fields, such as monitoring open-pit coal mines. Therefore, the main content and conclusions of this paper are as follows: (1) A method for Automatic Control Points Extraction from ICESat-2/ATL08 products was developed, integrating Land cover types and Phenological information (ACPELP), achieving a mean absolute error (MAE) of 1.05 m in the horizontal direction and 1.99 m in the vertical direction for stereo change measurements. This method helps correct image positioning errors, enabling the acquisition of geospatially aligned GaoFen-7 (GF-7) imagery. (2) A function-based classification system for open-pit coal mines was established, enabling precise extraction of stereoscopic change region to support accurate volumetric calculations. (3) A method for calculating the mining and stripping volume of open-pit coal mines based on GF-7 imagery is proposed. The method utilizes photogrammetry to extract elevation features and combines spectral features with elevation data to estimate stripping volumes, achieving an excellent error rate (ER) of 0.26%. The results indicate that our method is cost-effective and highly practical, filling the gap in accurate and comprehensive monitoring of land surface changes. Full article

Woody plants serve as crucial ecological barriers surrounding oases in arid and semi-arid regions, playing a vital role in maintaining the stability and supporting sustainable development of oases. However, their sparse distribution makes significant challenges in accurately mapping their spatial extent using medium-resolution remote sensing imagery. In this study, we utilized high-resolution Gaofen (GF-2) and Landsat 5/7/8 satellite images to quantify the relationship between vegetation growth and groundwater table depths (GTD) in a typical inland river basin from 1988 to 2021. Our findings are as follows: (1) Based on the D-LinkNet model, the distribution of woody plants was accurately extracted with an overall accuracy (OA) of 96.06%. (2) Approximately 95.33% of the desert areas had fractional woody plant coverage (FWC) values of less than 10%. (3) The difference between fractional woody plant coverage and fractional vegetation cover proved to be a fine indicator for delineating the range of desert-oasis ecotone. (4) The optimal GTD for Haloxylon ammodendron and Tamarix ramosissima was determined to be 5.51 m and 3.36 m, respectively. Understanding the relationship between woody plant growth and GTD is essential for effective ecological conservation and water resource management in arid and semi-arid regions. Full article

In forest ecosystems, the application of hyperspectral (HS) imagery offers unprecedented opportunities for refined identification and classification. The diversity and complexity of forest cover make it challenging for traditional remote-sensing techniques to capture subtle spectral differences. Hyperspectral imagery, however, can reveal the nuanced changes in different tree species, vegetation health status, and soil composition through its nearly continuous spectral information. This detailed spectral information is crucial for the monitoring, management, and conservation of forest resources. While Convolutional Neural Networks (CNNs) have demonstrated excellent local context modeling capabilities in HS image classification, their inherent network architecture limits the exploration and representation of spectral feature sequence properties. To address this issue, we have rethought HS image classification from a sequential perspective and proposed a hybrid model, the Spectral Weaver, which combines CNNs and Transformers. The Spectral Weaver replaces the traditional Multi-Head Attention mechanism with a Channel Attention mechanism (MCA) and introduces Centre-Differential Convolutional Layers (Conv2d-cd) to enhance spatial feature extraction capabilities. Additionally, we designed a cross-layer skip connection that adaptively learns to fuse “soft” residuals, transferring memory-like components from shallow to deep layers. Notably, the proposed model is a highly flexible backbone network, adaptable to both hyperspectral and multispectral image inputs. In comparison to traditional Visual Transformers (ViT), the Spectral Weaver innovates in several ways: (1) It introduces the MCA mechanism to enhance the mining of spectral feature sequence properties; (2) It employs Centre-Differential Convolutional Layers to strengthen spatial feature extraction; (3) It designs cross-layer skip connections to reduce information loss; (4) It supports both multispectral and hyperspectral inputs, increasing the model’s flexibility and applicability. By integrating global and local features, our model significantly improves the performance of HS image classification. We have conducted extensive experiments on the Gaofen dataset, multispectral data, and multiple hyperspectral datasets, validating the superiority of the Spectral Weaver model in forest hyperspectral image classification. The experimental results show that our model achieves 98.59% accuracy on multispectral data, surpassing ViT’s 96.30%. On the Jilin-1 dataset, our proposed algorithm achieved an accuracy of 98.95%, which is 2.17% higher than ViT. The model significantly outperforms classic ViT and other state-of-the-art backbone networks in classification performance. Not only does it effectively capture the spectral features of forest vegetation, but it also significantly improves the accuracy and robustness of classification, providing strong technical support for the refined management and conservation of forest resources. Full article

Transformer-based methods have demonstrated impressive performance in image super-resolution tasks. However, when applied to large-scale Earth observation images, the existing transformers encounter two significant challenges: (1) insufficient consideration of spatial correlation between adjacent ground objects; and (2) performance bottlenecks due to the underutilization of the upsample module. To address these issues, we propose a novel distance-enhanced strip attention transformer (DESAT). The DESAT integrates distance priors, easily obtainable from remote sensing images, into the strip window self-attention mechanism to capture spatial correlations more effectively. To further enhance the transfer of deep features into high-resolution outputs, we designed an attention-enhanced upsample block, which combines the pixel shuffle layer with an attention-based upsample branch implemented through the overlapping window self-attention mechanism. Additionally, to better simulate real-world scenarios, we constructed a new cross-sensor super-resolution dataset using Gaofen-6 satellite imagery. Extensive experiments on both simulated and real-world remote sensing datasets demonstrate that the DESAT outperforms state-of-the-art models by up to 1.17 dB along with superior qualitative results. Furthermore, the DESAT achieves more competitive performance in real-world tasks, effectively balancing spatial detail reconstruction and spectral transform, making it highly suitable for practical remote sensing super-resolution applications. Full article

As one of the major global food crops, the monitoring and management of the winter wheat planting area is of great significance for agricultural production and food security worldwide. Today, the development of high-resolution remote sensing imaging technology has provided rich sources of data for extracting the visual planting information of winter wheat. However, the existing research mostly focuses on extracting the planting plots that have a simple terrain structure. In the face of diverse terrain features combining mountainous areas, plains, and saline alkali land, as well as small-scale but complex planting structures, the extraction of planting plots through remote sensing imaging is subjected to great challenges in terms of recognition accuracy and model complexity. In this paper, we propose a modified Segformer model for extracting winter wheat planting plots with complex structures in rural areas based on the 0.8 m high-resolution multispectral data obtained from the Gaofen-2 satellite, which significantly improves the extraction accuracy and efficiency under complex conditions. In the encoder and decoder of this method, new modules were developed for the purpose of optimizing the feature extraction and fusion process. Specifically, the improvement measures of the proposed method include: (1) The MixFFN module in the original Segformer model is replaced with the Multi-Scale Feature Fusion Fully-connected Network (MSF-FFN) module, which enhances the model’s representation ability in handling complex terrain features through multi-scale feature extraction and position embedding convolution; furthermore, the DropPath mechanism is introduced to reduce the possibility of overfitting while improving the model’s generalization ability. (2) In the decoder part, after fusing features at four different scales, a CoordAttention module is added, which can precisely locate important regions with enhanced features in the images by utilizing the coordinate attention mechanism, therefore further improving the model’s extraction accuracy. (3) The model’s input data are strengthened by incorporating multispectral indices, which are also conducive to the improvement of the overall extraction accuracy. The experimental results show that the accuracy rate of the modified Segformer model in extracting winter wheat planting plots is significantly increased compared to traditional segmentation models, with the mean Intersection over Union (mIOU) and mean Pixel Accuracy (mPA) reaching 89.88% and 94.67%, respectively (an increase of 1.93 and 1.23 percentage points, respectively, compared to the baseline model). Meanwhile, the parameter count and computational complexity are significantly reduced compared to other similar models. Furthermore, when multispectral indices are input into the model, the mIOU and mPA reach 90.97% and 95.16%, respectively (an increase of 3.02 and 1.72 percentage points, respectively, compared to the baseline model). Full article

The coastal zone is abundant in natural resources but has become increasingly fragile in recent years due to climate change and extensive, improper exploitation. Accurate land use and land cover (LULC) mapping of coastal zones using remotely sensed data is crucial for monitoring environmental changes. Traditional classification methods based on statistical learning require significant spectral differences between ground objects. However, state-of-the-art end-to-end deep learning methods can extract advanced features from remotely sensed data. In this study, we employed ResNet50 as the feature extraction network within the U-Net architecture to achieve accurate classification of coastal areas and assess the model’s performance. Experiments were conducted using Gaofen-2 (GF-2) high-resolution remote sensing data from Shuangyue Bay, a typical coastal area in Guangdong Province. We compared the classification results with those obtained from two popular deep learning models, SegNet and DeepLab v3+, as well as two advanced statistical learning models, Support Vector Machine (SVM) and Random Forest (RF). Additionally, this study further explored the significance of Gray Level Co-occurrence Matrix (GLCM) texture features, Histogram Contrast (HC) features, and Normalized Difference Vegetation Index (NDVI) features in the classification of coastal areas. The research findings indicated that under complex ground conditions, the U-Net model achieved the highest overall accuracy of 86.32% using only spectral channels from GF-2 remotely sensed data. When incorporating multiple features, including spectrum, texture, contrast, and vegetation index, the classification accuracy of the U-Net algorithm significantly improved to 93.65%. The major contributions of this study are twofold: (1) it demonstrates the advantages of deep learning approaches, particularly the U-Net model, for LULC classification in coastal zones using high-resolution remote sensing images, and (2) it analyzes the contributions of spectral and spatial features of GF-2 data for different land cover types through a spectral and spatial combination method. Full article

Under the background of global change, the lake water environment is facing a huge threat from eutrophication. The rapid increase in curly-leaf pondweed (Potamogeton crispus L.) in recent years has seriously threatened the ecological balance and the water diversion safety of the eastern route of China’s South-to-North Water Diversion Project. The monitoring and control of curly-leaf pondweed is imperative in shallow lakes of northern China. Unmanned Aerial Vehicles (UAVs) have great potential for monitoring aquatic vegetation. However, merely using satellite remote sensing to detect submerged vegetation is not sufficient, and the monitoring of UAVs on aquatic vegetation is rarely systematically evaluated. In this study, taking Nansi Lake as a case, we employed Red–Green–Blue (RGB) UAV and satellite datasets to evaluate the monitoring of RGB Vegetation Indices (VIs) in pondweed and mapped the dynamic patterns of the pondweed Fractional Vegetation Coverage (FVC) in Nansi Lake. The pondweed FVC values were extracted using the RGB VIs and the machine learning method. The extraction of the UAV RGB images was evaluated by correlations, accuracy assessments and separability. The correlation between VIs and FVC was used to invert the pondweed FVC in Nansi Lake. The RGB VIs were also calculated using Gaofen-2 (GF-2) and were compared with UAV and Sentinel-2 data. Our results showed the following: (1) The RGB UAV could effectively monitor the FVC of pondweed, especially when using Support Vector Machine that (SVM) has a high ability to recognize pondweed in UAV RGB images. Two RGB VIs, RCC and RGRI, appeared best suited for monitoring aquatic plants. The correlations between four RGB VIs based on GF-2, i.e., GCC, BRI, VDVI, and RGBVI and FVC_SVM calculated by the UAV (p < 0.01) were better than those obtained with other RGB VIs. Thus, the RGB VIs of GF-2 were not as effective as those of the UAV in pondweed monitoring. (2) The binomial estimation model constructed by the Normalized Difference Water Index (NDWI) of Sentinel-2 showed a high accuracy (R² = 0.7505, RMSE = 0.169) for pondweed FVC and can be used for mapping the FVC of pondweed in Nansi Lake. (3) Combined with the Sentinel-2 time-series data, we mapped the dynamic patterns of pondweed FVC in Nansi Lake. It was determined that the flooding of pondweed in Nansi Lake has been alleviated in recent years, but the rapid increase in pondweed in part of Nansi Lake remains a challenging management issue. This study provides practical tools and methodology for the innovative remote sensing monitoring of submerged vegetation. Full article

The development of green spaces in urban parks can significantly enhance the quality of the urban and ecological environment. This paper utilizes 2021 Gaofen-7 (GF-7) satellite remote sensing images as its primary data source and uses deep learning algorithms for the precise extraction of the green space coverage within Beijing’s fifth ring road. It also incorporates the park points of interest (POI) information, road data, and other auxiliary data to extract green park space details. The analysis focuses on examining the relationship between supply and demand in the spatial allocation of green park spaces from an accessibility perspective. The main findings are as follows: (1) The application of deep learning algorithms improves the accuracy of green space extraction by 10.68% compared to conventional machine methods. (2) The distribution of parks and green spaces within the fifth ring road of Beijing is uneven, showing a clear pattern of “more in the north and less in the south”. The accessibility within a five-minute service radius achieves a coverage rate of 46.65%, with a discernible blind zone in the southeast. (3) There is an imbalance in the per capita green space location entropy within the fifth ring road of Beijing, there is a big difference in per capita green space location entropy (44.19), and social fairness needs to be improved. The study’s outcomes unveil the intricate relationship between service capacity and spatial allocation, shedding light on the supply and demand dynamics of parks and green spaces within Beijing’s fifth ring road. This insight will contribute to the construction of ecologically sustainable and aesthetically pleasing living spaces in modern megacities. Full article

A panchromatic and multispectral sensor (PMS) and a wide-field-of-view (WFV) sensor were fitted aboard the Gaofen6 (GF6) satellite, which was launched on 2 June 2018. This study used the Landsat9-Operational Land Imager 2 and Sentinel2-Multispectral Instrument as reference sensors to perform radiometric cross-calibration on GF6-PMS and WFV data at the Dunhuang calibration site. The four selected sensor images were all acquired on the same day. The results indicate that: the calibration results between different reference sensors can be controlled within 3%, with the maximum difference from the official coefficients being 8.78%. A significant difference was observed between the coefficients obtained by different reference sensors when spectral band adjustment factor (SBAF) correction was not performed; from the two sets of validation results, the maximum mean relative difference in the near-infrared band was 9.46%, with the WFV sensor showing better validation results. The validation of calibration coefficients based on synchronous ground observation data and the analysis of the impact of different SBAF methods on the calibration results indicated that Landsat9 is more suitable as a reference sensor for radiometric cross-calibration of GF6-PMS and WFV. Full article