Remote Sensing

20 pages, 31614 KiB

Open AccessArticle

Fine-Scale Classification of Dominant Vegetation Communities in Coastal Wetlands Using Color-Enhanced Aerial Images

by Yixian Liu, Yiheng Zhang, Xin Zhang, Chunguang Che, Chong Huang, He Li, Yu Peng, Zishen Li and Qingsheng Liu

Remote Sens. 2025, 17(16), 2848; https://doi.org/10.3390/rs17162848 - 15 Aug 2025

Abstract

Monitoring salt marsh vegetation in the Yellow River Delta (YRD) wetland is the basis of wetland research, which is of great significance for the further protection and restoration of wetland ecological functions. In the existing remote sensing technologies for wetland salt marsh vegetation [...] Read more.

Monitoring salt marsh vegetation in the Yellow River Delta (YRD) wetland is the basis of wetland research, which is of great significance for the further protection and restoration of wetland ecological functions. In the existing remote sensing technologies for wetland salt marsh vegetation classification, the object-oriented classification method effectively produces landscape patches similar to wetland vegetation and improves the spatial consistency and accuracy of the classification. However, the vegetation classes of the YRD are mixed with uneven distribution, irregular texture, and significant color variation. In order to solve the problem, this study proposes a fine-scale classification of dominant vegetation communities using color-enhanced aerial images. The color information is used to extract the color features of the image. Various features including spectral features, texture features and vegetation features are extracted from the image objects and used as inputs for four machine learning classifiers: random forest (RF), support vector machine (SVM), k-nearest neighbor (KNN) and maximum likelihood (MLC). The results showed that the accuracy of the four classifiers in classifying vegetation communities was significantly improved by adding color features. RF had the highest OA and Kappa coefficients of 96.69% and 0.9603. This shows that the classification method based on color enhancement can effectively distinguish between vegetation and non-vegetation and extract each vegetation type, which provides an effective technical route for wetland vegetation classification in aerial imagery. Full article

(This article belongs to the Special Issue Remote Sensing in Coastal Vegetation Monitoring)

► Show Figures

Figure 1

47 pages, 10086 KiB

Open AccessArticle

Tree Type Classification from ALS Data: A Comparative Analysis of 1D, 2D, and 3D Representations Using ML and DL Models

by Sead Mustafić, Mathias Schardt and Roland Perko

Remote Sens. 2025, 17(16), 2847; https://doi.org/10.3390/rs17162847 - 15 Aug 2025

Abstract

Accurate classification of individual tree types is a key component in forest inventory, biodiversity monitoring, and ecological modeling. This study evaluates and compares multiple Machine Learning (ML) and Deep Learning (DL) approaches for tree type classification based on Airborne Laser Scanning (ALS) data. [...] Read more.

Accurate classification of individual tree types is a key component in forest inventory, biodiversity monitoring, and ecological modeling. This study evaluates and compares multiple Machine Learning (ML) and Deep Learning (DL) approaches for tree type classification based on Airborne Laser Scanning (ALS) data. A mixed-species forest in southeastern Austria, Europe, served as the test site, with spruce, pine, and a grouped class of broadleaf species as target categories. To examine the impact of data representation, ALS point clouds were transformed into four distinct structures: 1D feature vectors, 2D raster profiles, 3D voxel grids, and unstructured 3D point clouds. A comprehensive dataset, combining field measurements and manually annotated aerial data, was used to train and validate 45 ML and DL models. Results show that DL models based on 3D point clouds achieved the highest overall accuracy (up to 88.1%), followed by multi-view 2D raster and voxel-based methods. Traditional ML models performed well on 1D data but struggled with high-dimensional inputs. Spruce trees were classified most reliably, while confusion between pine and broadleaf species remained challenging across methods. The study highlights the importance of selecting suitable data structures and model types for operational tree classification and outlines potential directions for improving accuracy through multimodal and temporal data fusion. Full article

(This article belongs to the Section Forest Remote Sensing)

21 pages, 6984 KiB

Open AccessArticle

Limitations of Polar-Orbiting Satellite Observations inCapturing the Diurnal Variability of Tropospheric NO₂: A Case Study Using TROPOMI, GOME-2C, and Pandora Data

by Yichen Li, Chao Yu, Jing Fan, Meng Fan, Ying Zhang, Jinhua Tao and Liangfu Chen

Remote Sens. 2025, 17(16), 2846; https://doi.org/10.3390/rs17162846 - 15 Aug 2025

Abstract

Nitrogen dioxide (NO₂) plays a crucial role in environmental processes and public health. In recent years, NO₂ pollution has been monitored using a combination of in situ measurements and satellite remote sensing, supported by the development of advanced retrieval algorithms. [...] Read more.

Nitrogen dioxide (NO₂) plays a crucial role in environmental processes and public health. In recent years, NO₂ pollution has been monitored using a combination of in situ measurements and satellite remote sensing, supported by the development of advanced retrieval algorithms. With advancements in satellite technology, large-scale NO₂ monitoring is now feasible through instruments such as GOME-2C and TROPOMI. However, the fixed local overpass times of polar-orbiting satellites limit their ability to capture the complete diurnal cycle of NO₂, introducing uncertainties in emission estimation and pollution trend analysis. In this study, we evaluated differences in NO₂ observations between GOME-2C (morning overpass at ~09:30 LT) and TROPOMI (afternoon overpass at ~13:30 LT) across three representative regions—East Asia, Central Africa, and Europe—that exhibit distinct emission sources and atmospheric conditions. By comparing satellite-derived tropospheric NO₂ column densities with ground-based measurements from the Pandora network, we analyzed spatial distribution patterns and seasonal variability in NO₂ concentrations. Our results show that East Asia experiences the highest NO₂ concentrations in densely populated urban and industrial areas. During winter, lower boundary layer heights and weakened photolysis processes lead to stronger accumulation of NO₂ in the morning. In Central Africa, where biomass burning is the dominant emission source, afternoon fire activity is significantly higher, resulting in a substantial difference (1.01 × 10¹⁶ molecules/cm²) between GOME-2C and TROPOMI observations. Over Europe, NO₂ pollution is primarily concentrated in Western Europe and along the Mediterranean coast, with seasonal peaks in winter. In high-latitude regions, weaker solar radiation limits the photochemical removal of NO₂, causing concentrations to continue rising into the afternoon. These findings demonstrate that differences in polar-orbiting satellite overpass times can significantly affect the interpretation of daily NO₂ variability, especially in regions with strong diurnal emissions or meteorological patterns. This study highlights the observational limitations of fixed-time satellites and offers an important reference for the future development of geostationary satellite missions, contributing to improved strategies for NO₂ pollution monitoring and control. Full article

(This article belongs to the Special Issue Advanced Remote Sensing Approaches for Multi-Scale Atmospheric Components Monitoring and Impact Assessment)

27 pages, 8994 KiB

Open AccessArticle

Lane Graph Extraction from Aerial Imagery via Lane Segmentation Refinement with Diffusion Models

by Antonio Ruiz, Andrew Melnik, Nicolo Savioli, Dong Wang, Yanfeng Zhang and Helge Ritter

Remote Sens. 2025, 17(16), 2845; https://doi.org/10.3390/rs17162845 - 15 Aug 2025

Abstract

The lane graph is critical for applications such as autonomous driving and lane-level route planning. While previous research has focused on extracting lane-level graphs from aerial imagery using convolutional neural networks (CNNs) followed by post-processing segmentation-to-graph algorithms, these methods often face challenges in [...] Read more.

The lane graph is critical for applications such as autonomous driving and lane-level route planning. While previous research has focused on extracting lane-level graphs from aerial imagery using convolutional neural networks (CNNs) followed by post-processing segmentation-to-graph algorithms, these methods often face challenges in producing sharp and complete segmentation masks. Challenges such as occlusions, variations in lighting, and changes in road texture can lead to incomplete and inaccurate lane masks, resulting in poor-quality lane graphs. To address these challenges, we propose a novel approach that refines the lane masks, output by a CNN, using diffusion models. Experimental results on a publicly available dataset demonstrate that our method outperforms existing methods based solely on CNNs or diffusion models, particularly in terms of graph connectivity. Our lane mask refinement approach enhances the quality of the extracted lane graph, yielding gains of approximately 1.5% in GEO F1 and 3.5% in TOPO F1 scores over the best-performing CNN-based method, and improvements of 28% and 34%, respectively, compared to a prior diffusion-based approach. Both GEO F1 and TOPO F1 scores are critical metrics for evaluating lane graph quality. Additionally, ablation studies are conducted to evaluate the individual components of our approach, providing insights into their respective contributions and effectiveness. Full article

(This article belongs to the Special Issue AI-Driven Mapping Using Remote Sensing Data)

► Show Figures

Figure 1

31 pages, 2290 KiB

Open AccessArticle

On the Space Observation of Resident Space Objects (RSOs) in Low Earth Orbits (LEOs)

by Angel Porras-Hermoso, Randa Qashoa, Regina S. K. Lee, Javier Cubas and Santiago Pindado

Remote Sens. 2025, 17(16), 2844; https://doi.org/10.3390/rs17162844 - 15 Aug 2025

Abstract

Space debris is an increasingly severe problem in the space industry. According to projections, the number of satellites will increase from the current 10,000 to 100,000 by 2030, specially in LEO orbits. This significant rise in the number of satellites threatens space sustainability, [...] Read more.

Space debris is an increasingly severe problem in the space industry. According to projections, the number of satellites will increase from the current 10,000 to 100,000 by 2030, specially in LEO orbits. This significant rise in the number of satellites threatens space sustainability, forcing satellites to perform more maneuvers to avoid impacts or leading to the production of more and more space debris due to collisions (Kessler Syndrome). Consequently, substantial efforts have been made to detect and track space debris, leading to the development of the current catalogs. However, with existing technology, detecting and tracking small debris remains challenging. In order to improve the current system, several proposals of Space-Based Situational Awareness (SBSA) have been made. These proposals involve satellites equipped with telescopes to detect space debris and determine their orbits. Unlike prior works, focused primarily on detection rates, this research aims to quantify their accuracy in orbit determination as a function of observation duration, the number of observers, and sensor precision. The Unscented Kalman Filter (UKF) is employed as the core estimation algorithm, leveraging both simulated single-case analyses and Monte Carlo simulations to evaluate system performance under various configurations and uncertainties. The results indicate that a constellation of at least three observers with high-precision instruments and sub-kilometer positioning accuracy can reliably estimate debris orbits within an observation period of 4–7 min, with the mean error in position and velocity obtained being 2.2–3 km and 3–4 m/s, respectively. These findings offer critical insights for designing future SBSA constellations and optimizing their operational parameters to address the growing challenge of orbital debris. Full article

28 pages, 5112 KiB

Open AccessArticle

Remote Sensing and Machine Learning Uncover Dominant Drivers of Carbon Sink Dynamics in Subtropical Mountain Ecosystems

by Leyan Xia, Hongjian Tan, Jialong Zhang, Kun Yang, Chengkai Teng, Kai Huang, Jingwen Yang and Tao Cheng

Remote Sens. 2025, 17(16), 2843; https://doi.org/10.3390/rs17162843 - 15 Aug 2025

Abstract

Net ecosystem productivity (NEP) serves as a key indicator for assessing regional carbon sink potential, with its dynamics regulated by nonlinear interactions among multiple factors. However, its driving factors and their coupling processes remain insufficiently characterized. This study investigated terrestrial ecosystems in Yunnan [...] Read more.

Net ecosystem productivity (NEP) serves as a key indicator for assessing regional carbon sink potential, with its dynamics regulated by nonlinear interactions among multiple factors. However, its driving factors and their coupling processes remain insufficiently characterized. This study investigated terrestrial ecosystems in Yunnan Province, China, to elucidate the drivers of NEP using 14 environmental factors (including topography, meteorology, soil texture, and human activities) and 21 remote sensing features. We developed a research framework based on “Feature Selection–Machine Learning–Mechanism Interpretation.” The results demonstrated that the Variable Selection Using Random Forests (VSURF) feature selection method effectively reduced model complexity. The selected features achieved high estimation accuracy across three machine learning models, with the eXtreme Gradient Boosting Regression (XGBR) model performing optimally (R² = 0.94, RMSE = 76.82 gC/(m²·a), MAE = 55.11 gC/(m²·a)). Interpretation analysis using the SHAP (SHapley Additive exPlanations) method revealed the following: (1) The Enhanced Vegetation Index (EVI), soil pH, solar radiation, air temperature, clay content, precipitation, sand content, and vegetation type were the primary drivers of NEP in Yunnan. Notably, EVI’s importance exceeded that of other factors by approximately 3 to 10 times. (2) Significant interactions existed between soil texture and temperature: Under low-temperature conditions (−5 °C to 12.15 °C), moderate clay content (13–25%) combined with high sand content (40–55%) suppressed NEP. Conversely, within the medium to high temperature range (5 °C to 23.79 °C), high clay content (25–40%) coupled with low sand content (25–43%) enhanced NEP. These findings elucidate the complex driving mechanisms of NEP in subtropical ecosystems, confirming the dominant role of EVI in carbon sequestration and revealing nonlinear regulatory patterns in soil–temperature interactions. This study provides not only a robust “Feature Selection–Machine Learning–Mechanism Interpretation” modeling framework for assessing carbon budgets in mountainous regions but also a scientific basis for formulating regional carbon management policies. Full article

(This article belongs to the Section Ecological Remote Sensing)

25 pages, 6271 KiB

Open AccessArticle

UAV-LiDAR-Based Study on AGB Response to Stand Structure and Its Estimation in Cunninghamia Lanceolata Plantations

by Yuqi Cao, Yinyin Zhao, Jiuen Xu, Qing Fang, Jie Xuan, Lei Huang, Xuejian Li, Fangjie Mao, Yusen Sun and Huaqiang Du

Remote Sens. 2025, 17(16), 2842; https://doi.org/10.3390/rs17162842 - 15 Aug 2025

Abstract

Forest spatial structure is of significant importance for studying forest biomass accumulation and management. However, above-ground biomass (AGB) estimation based on satellite remote sensing struggles to capture forest spatial structure information, which to some extent affects the accuracy of AGB estimation. To address [...] Read more.

Forest spatial structure is of significant importance for studying forest biomass accumulation and management. However, above-ground biomass (AGB) estimation based on satellite remote sensing struggles to capture forest spatial structure information, which to some extent affects the accuracy of AGB estimation. To address this issue, this study focused on Chinese fir (Cunninghamia lanceolata) plantations in Zhejiang Province. Using UAV-LiDAR (unmanned aerial vehicle light detection and ranging) data and a seed-point-based individual tree segmentation algorithm, information on individual fir trees was obtained. Building on this foundation, structural parameters such as neighborhood comparison (U), crowding degree (C), uniform angle index (W), competition index (CI), and canopy openness (K) were calculated, and their distribution characteristics analyzed. Finally, these parameters were integrated with UAV-LiDAR point cloud features to build machine learning models, and a geographical detector was used to quantify their contribution to AGB estimation. The research findings indicate the following: (1) The studied stands exhibited a random spatial pattern, moderate competition, and sufficient growing space. (2) A significant correlation existed between the U and AGB (r > 0.6), followed by CI. The optimal stand structure for AGB accumulation was C = 0.25, U < 0.5, CI in (0, 0.8], and K > 0.3. (3) The four machine learning models constructed by coupling spatial structure with point cloud features all improved the accuracy of AGB estimation for the fir forest to some extent. Among them, the XGBoost model performed best, achieving a model accuracy (R²) of 0.92 and a relatively low error (RMSE = 14.02 kg). (4) Geographical detector analysis indicated that U and CI contributed most to AGB estimation, with q-values of 0.44 and 0.37, respectively. Full article

► Show Figures

Figure 1

22 pages, 76137 KiB

Open AccessArticle

CS-FSDet: A Few-Shot SAR Target Detection Method for Cross-Sensor Scenarios

by Changzhi Liu, Yibin He, Xiuhua Zhang, Yanwei Wang, Zhenyu Dong and Hanyu Hong

Remote Sens. 2025, 17(16), 2841; https://doi.org/10.3390/rs17162841 - 15 Aug 2025

Abstract

Synthetic Aperture Radar (SAR) plays a pivotal role in remote-sensing target detection. However, domain shift caused by distribution discrepancies across sensors, coupled with the scarcity of target-domain samples, severely restricts the generalization and practical performance of SAR detectors. To address these challenges, this [...] Read more.

Synthetic Aperture Radar (SAR) plays a pivotal role in remote-sensing target detection. However, domain shift caused by distribution discrepancies across sensors, coupled with the scarcity of target-domain samples, severely restricts the generalization and practical performance of SAR detectors. To address these challenges, this paper proposes a few-shot SAR target-detection framework tailored for cross-sensor scenarios (CS-FSDet), enabling efficient transfer of source-domain knowledge to the target domain. First, to mitigate inter-domain feature-distribution mismatch, we introduce a Multi-scale Uncertainty-aware Bayesian Distribution Alignment (MUBDA) strategy. By modeling features as Gaussian distributions with uncertainty and performing dynamic weighting based on uncertainty, MUBDA achieves fine-grained distribution-level alignment of SAR features under different resolutions. Furthermore, we design an Adaptive Cross-domain Interactive Coordinate Attention (ACICA) module that computes cross-domain spatial-attention similarity and learns interaction weights adaptively, thereby suppressing domain-specific interference and enhancing the expressiveness of domain-shared target features. Extensive experiments on two cross-sensor few-shot detection tasks, HRSID→SSDD and SSDD→HRSID, demonstrate that the proposed method consistently surpasses state-of-the-art approaches in mean Average Precision (mAP) under 1-shot to 10-shot settings. Full article

(This article belongs to the Special Issue SAR Target Detection and Recognition with Intelligent Methods and Its Applications)

► Show Figures

Figure 1

21 pages, 56969 KiB

Open AccessArticle

Exploring Impacts of Land Use and Cover Changes on Ecosystem Services on the Qinghai-Xizang Plateau

by Yingxin Li, Zhiding Hu, Chenli Liu, Xin Yang, Zhe Zhang, Weizhao Sun, Fuchang Niu, Enwei Zhang and Qike Yang

Remote Sens. 2025, 17(16), 2840; https://doi.org/10.3390/rs17162840 - 15 Aug 2025

Abstract

The Qinghai-Xizang Plateau (QXP), as Asia’s “Water Tower” and a global climate regulation hub, provides essential ecosystem services that sustain global ecological security and the well-being of 2 billion people. However, the fine-scale relationship between land use changes and ecosystem services on the [...] Read more.

The Qinghai-Xizang Plateau (QXP), as Asia’s “Water Tower” and a global climate regulation hub, provides essential ecosystem services that sustain global ecological security and the well-being of 2 billion people. However, the fine-scale relationship between land use changes and ecosystem services on the QXP, as well as the underlying driving mechanisms, remain under researched. Therefore, the InVEST model was used in this study to assess the distribution of key factors of ecosystem services—habitat quality (HQ). A four-quadrant diagram method was proposed to quantitatively identify the dynamic relationship between land use changes and habitat quality. The XGBoost-SHAP algorithm was used to determine the main driving factors affecting the conversion between coordinated and conflicting areas. This study finds the following: (1) the proportion of land area undergoing land use change on the QXP is approximately 7.5%, with significant grassland degradation; (2) habitat quality distribution shows a high edge-low northern pattern, and forest land has the greatest impact on habitat quality changes; (3) in the past 20 years, the relationship between land use changes and habitat quality has shown a trend of coordinated development, with a 50:1 ratio of coordination to conflict evolution. The relationship between cities and water source areas has experienced the most dramatic changes; moreover, in the conversion between coordination and conflict regions, natural factors are the main drivers, followed by socio-cultural factors. This study reveals the dynamic relationship between land use changes and habitat quality interactions on the QXP, providing a scientific basis for regional sustainable development and related planning. Full article

► Show Figures

Figure 1

23 pages, 13218 KiB

Open AccessArticle

Driving Analyses of the Effects of Climate Change and Human Activity on the Ecological Environmental Quality of the North China Plain

by Zefeng Wei, Shuting Wang, Yunlan Guan, Yuecan Hu, Siyao Wang and Li Shen

Remote Sens. 2025, 17(16), 2839; https://doi.org/10.3390/rs17162839 - 15 Aug 2025

Abstract

Understanding the dynamic changes in the quality of the ecological environment and its potential driving forces is essential for protecting regional ecosystems and promoting sustainable development. In this study, we developed an improved remote sensing ecological index (IRSEI) by integrating the kernel normalized [...] Read more.

Understanding the dynamic changes in the quality of the ecological environment and its potential driving forces is essential for protecting regional ecosystems and promoting sustainable development. In this study, we developed an improved remote sensing ecological index (IRSEI) by integrating the kernel normalized difference vegetation index (kNDVI) with an abundance index (AI) and conducted a comprehensive analysis of the spatiotemporal evolution of the quality of the ecological environment in the North China Plain (NCP) from 2000 to 2020. A multistep driving analysis framework was established to identify key climatic factors via the XGBoost algorithm and to quantify the effects of climate change and human activities through partial correlation analysis and a multiple regression residual model. The results indicate the following: (1) From 2000 to 2020, the ecological quality of the NCP significantly improved, with the average IRSEI increasing from 0.41 to 0.45. The proportion of areas with “good” or “excellent” ecological quality increased, revealing a south–north gradient, with higher values in the southern part and lower values in the northern part of the NCP. (2) Among the key climatic variables, surface temperature was significantly negatively correlated with the IRSEI, whereas atmospheric pressure and evapotranspiration were significantly positively correlated. (3) Approximately 51.97% of the ecological quality changes were jointly driven by climate change and human activities, with the contribution of human activities (28.80%) exceeding that of climate change (19.23%). These findings provide a scientific basis for understanding the driving mechanisms behind ecological environment changes and support ecological restoration and coordinated human–environment development in the context of climate change. Full article

(This article belongs to the Special Issue Remote Sensing of Ecosystem Structure and Function Dynamics Due to Climate Change and Human Activities (2nd Edition))

► Show Figures

Figure 1

16 pages, 1072 KiB

Open AccessArticle

ω − k MUSIC Algorithm for Subsurface Target Localization

by Antonio Cuccaro, Angela Dell’Aversano, Maria Antonia Maisto, Rosa Scapaticci, Adriana Brancaccio and Raffaele Solimene

Remote Sens. 2025, 17(16), 2838; https://doi.org/10.3390/rs17162838 - 15 Aug 2025

Abstract

This paper addresses the problem of subsurface target localization from single-snapshot multimonostatic and multifrequency radar measurements. In this context, the use of subspace projection methods—known for their super-resolution capabilities—is hindered by the rank deficiency of the data correlation matrix and the lack of [...] Read more.

This paper addresses the problem of subsurface target localization from single-snapshot multimonostatic and multifrequency radar measurements. In this context, the use of subspace projection methods—known for their super-resolution capabilities—is hindered by the rank deficiency of the data correlation matrix and the lack of a Vandermonde structure, especially in near-field configurations and layered media. To overcome this issue, we propose a novel pre-processing strategy that transforms the measured data into the

ω - k

domain, thereby restoring the structural conditions required for subspace-based detection. The resulting algorithm, referred to as

ω - k

MUSIC, enables the application of subspace projection techniques in scenarios where traditional smoothing procedures are not viable. Numerical experiments in a 2-D scalar configuration demonstrate the effectiveness of the proposed method in terms of resolution and robustness under various noise conditions. A Monte Carlo simulation study is also included to provide a quantitative assessment of localization accuracy. Comparisons with conventional migration imaging highlight the superior performance of the proposed approach. Full article

► Show Figures

Figure 1

20 pages, 6431 KiB

Open AccessArticle

Characterizing Role of Spatial Features in Improving Mangrove Classification—A Case Study over the Mesoamerican Reef Region

by Suvarna M. Punalekar, A. Justin Nowakowski, Steven W. J. Canty, Craig Fergus, Qiongyu Huang, Melissa Songer and Grant M. Connette

Remote Sens. 2025, 17(16), 2837; https://doi.org/10.3390/rs17162837 - 15 Aug 2025

Abstract

Mangrove forests are among the world’s most vital coastal ecosystems. Mapping mangrove cover from local to global scales using spectral data and machine learning models is a well-established method. While non-spectral contextual datasets (spatial features) have also been incorporated into such models, the [...] Read more.

Mangrove forests are among the world’s most vital coastal ecosystems. Mapping mangrove cover from local to global scales using spectral data and machine learning models is a well-established method. While non-spectral contextual datasets (spatial features) have also been incorporated into such models, the contribution of these additional features to improving mangrove mapping remains underexplored. Using the Mesoamerican Reef Region as a case study, we evaluate the effectiveness of incorporating spatial features in binary mangrove classification to enhance mapping accuracy. We compared an aspatial model that includes only spectral data with three spatial models: two included features such as geographic coordinates, elevation, and proximity to coastlines and streams, while the third integrated a geostatistical approach using Inverse Distance Weighted (IDW) interpolation. Spectral inputs included bands and indices derived from Sentinel-1 and Sentinel-2, and all models were implemented using the Random Forest algorithm in Google Earth Engine. Results show that spatial features reduced omission errors without increasing commission errors, enhancing the model’s ability to capture spatial variability. Models using geographic coordinates and elevation performed comparably to those with additional environmental variables, with storm frequency and distance to streams emerging as important predictors in the Mesoamerican Reef region. In contrast, the IDW-based model underperformed, likely due to overfitting and limited representation of local spectral variation. Spatial analyses show that models incorporating spatial features produced more continuous mangrove patches and removed some false positives in non-mangrove areas. These findings highlight the value of spatial features in improving classification accuracy, especially in regions with ecologically diverse mangroves across varied environments. By integrating spatial context, these models support more accurate, locally relevant mangrove maps that are essential for effective conservation and management. Full article

(This article belongs to the Special Issue Remote Sensing in Mangroves IV)

► Show Figures

Figure 1

20 pages, 7578 KiB

Open AccessArticle

Cross Attention Based Dual-Modality Collaboration for Hyperspectral Image and LiDAR Data Classification

by Khanzada Muzammil Hussain, Keyun Zhao, Yang Zhou, Aamir Ali and Ying Li

Remote Sens. 2025, 17(16), 2836; https://doi.org/10.3390/rs17162836 - 15 Aug 2025

Abstract

Advancements in satellite sensor technology have enabled access to diverse remote sensing (RS) data from multiple platforms. Hyperspectral Image (HSI) data offers rich spectral detail for material identification, while LiDAR captures high-resolution 3D structural information, making the two modalities naturally complementary. By fusing [...] Read more.

Advancements in satellite sensor technology have enabled access to diverse remote sensing (RS) data from multiple platforms. Hyperspectral Image (HSI) data offers rich spectral detail for material identification, while LiDAR captures high-resolution 3D structural information, making the two modalities naturally complementary. By fusing HSI and LiDAR, we can mitigate the limitations of each and improve tasks like land cover classification, vegetation analysis, and terrain mapping through more robust spectral–spatial feature representation. However, traditional multi-scale feature fusion models often struggle with aligning features effectively, which can lead to redundant outputs and diminished spatial clarity. To address these issues, we propose the Cross Attention Bridge for HSI and LiDAR (CAB-HL), a novel dual-path framework that employs a multi-stage cross-attention mechanism to guide the interaction between spectral and spatial features. In CAB-HL, features from each modality are refined across three progressive stages using cross-attention modules, which enhance contextual alignment while preserving the distinctive characteristics of each modality. These fused representations are subsequently integrated and passed through a lightweight classification head. Extensive experiments on three benchmark RS datasets demonstrate that CAB-HL consistently outperforms existing state-of-the-art models, confirm that CAB-HL consistently outperforms in learning deep joint representations for multimodal classification tasks. Full article

(This article belongs to the Special Issue Artificial Intelligence Remote Sensing for Earth Observation)

► Show Figures

Figure 1

21 pages, 6300 KiB

Open AccessArticle

Comparison of Machine Learning Algorithms for Simulating Brightness Temperature Using Data from the Tianjun Soil Moisture Observation Network

by Shaoning Lv, Zixi Liu and Jun Wen

Remote Sens. 2025, 17(16), 2835; https://doi.org/10.3390/rs17162835 - 15 Aug 2025

Abstract

The L-band radiative transfer-forward modeling plays a crucial role in data assimilation for meteorological forecasting. By utilizing information from the underlying surface (typically land surface parameters and variables), such as soil moisture, soil temperature, snow cover, freeze–thaw status, and vegetation, the corresponding brightness [...] Read more.

The L-band radiative transfer-forward modeling plays a crucial role in data assimilation for meteorological forecasting. By utilizing information from the underlying surface (typically land surface parameters and variables), such as soil moisture, soil temperature, snow cover, freeze–thaw status, and vegetation, the corresponding brightness temperatures can be simulated through the physical processes described by radiative transfer models. Data assimilation becomes meaningful when the errors introduced by the simulated brightness temperatures are smaller than the simulation accuracy of the land surface variables. However, radiative transfer models at the L-band cannot accurately simulate TB operationally. In this study, four machine learning methods, including random forest (RF), long short-term memory (LSTM), support vector machine (SVM), and deep neural networks (DNN), are employed to reconstruct the forward relationship from land surface parameters to brightness temperatures, serving as an alternative to traditional radiative transfer models. The performance of these methods is evaluated using ground-truthed soil moisture data, soil texture static data, and leaf area index (LAI). The results indicate that DNN and RF exhibit superior performance, with DNN achieving the lowest average unbiased root mean square error (ubRMSE) of 6.238 K for vertical polarization brightness temperature (TBv) and 9.033 K for horizontal polarization brightness temperature (TBh). Regarding correlation coefficients between the retrieved brightness temperatures and satellite measurements, RF leads for H-polarized TB with a value of 0.943, while both RF and SVM perform well for V-polarized TB with values of 0.930 and 0.932, respectively. In conclusion, our study shows that DNN is the optimal method for retrieving brightness temperatures, outperforming other machine learning approaches regarding error metrics and correlation with satellite measurements. These findings highlight the potential of DNN in improving data assimilation processes in meteorological forecasting. Full article

(This article belongs to the Special Issue Microwave Remote Sensing of Soil Moisture II)

► Show Figures

Figure 1

22 pages, 7227 KiB

Open AccessArticle

Mechanisms Driving Recent Sea-Level Acceleration in the Gulf of Guinea

by Ayinde Akeem Shola, Huaming Yu, Kejian Wu and Nir Krakauer

Remote Sens. 2025, 17(16), 2834; https://doi.org/10.3390/rs17162834 - 15 Aug 2025

Abstract

The Gulf of Guinea is undergoing accelerated sea-level rise (SLR), with localized rates surpassing 10 mm yr⁻¹, more than double the global mean. Integrating GRACE/FO ocean mass data, reanalysis products, and machine learning, we identify a regime shift in the regional [...] Read more.

The Gulf of Guinea is undergoing accelerated sea-level rise (SLR), with localized rates surpassing 10 mm yr⁻¹, more than double the global mean. Integrating GRACE/FO ocean mass data, reanalysis products, and machine learning, we identify a regime shift in the regional sea-level budget post-2015. Over 60% of observed SLR near major riverine outlets stems from ocean mass increase, driven primarily by intensified terrestrial hydrological discharge, marking a transition from steric to barystatic and manometric dominance. This shift coincides with enhanced monsoonal precipitation, wind-forced equatorial wave adjustments, and Atlantic–Pacific climate coupling. Piecewise regression reveals a significant 2015 breakpoint, with mean coastal SLR rates increasing from 2.93 ± 0.1 to 5.4 ± 0.25 mm yr⁻¹ between 1993 and 2014, and 2015 and 2023. GRACE data indicate extreme mass accumulation (>10 mm yr⁻¹) along the eastern Gulf coast, tied to elevated river discharge and estuarine retention. Dynamical analysis reveals the reorganization of wind field intensification, which modifies Rossby wave dispersion and amplifies zonal water mass convergence. Random forest modeling attributes 16% of extreme SLR variance to terrestrial runoff (comparable to wind stress at 19%), underscoring underestimated land–ocean interactions. Current climate models underrepresent manometric contributions by 20–45%, introducing critical projection biases for high-runoff regions. The societal implications are severe, with >400 km² of urban land in Lagos and Abidjan vulnerable to inundation by 2050. These findings reveal a hybrid steric–manometric regime in the Gulf of Guinea, challenging existing paradigms and suggesting analogous dynamics may operate across tropical margins. This calls for urgent model recalibration and tailored regional adaptation strategies. Full article

(This article belongs to the Special Issue Advanced Applications of Remote Sensing in Monitoring Marine Environment (Second Edition))

► Show Figures

Figure 1

27 pages, 9913 KiB

Open AccessArticle

BioLiteNet: A Biomimetic Lightweight Hyperspectral Image Classification Model

by Bo Zeng, Suwen Chao, Jialang Liu, Yanming Guo, Yingmei Wei, Huimin Yi, Bin Xie, Yaowen Hu and Lin Li

Remote Sens. 2025, 17(16), 2833; https://doi.org/10.3390/rs17162833 - 14 Aug 2025

Abstract

Hyperspectral imagery (HSI) has demonstrated significant potential in remote sensing applications because of its abundant spectral and spatial information. However, current mainstream hyperspectral image classification models are generally characterized by high computational complexity, structural intricacy, and a strong reliance on training samples, which [...] Read more.

Hyperspectral imagery (HSI) has demonstrated significant potential in remote sensing applications because of its abundant spectral and spatial information. However, current mainstream hyperspectral image classification models are generally characterized by high computational complexity, structural intricacy, and a strong reliance on training samples, which poses challenges in meeting application demands under resource-constrained conditions. To this end, a lightweight hyperspectral image classification model inspired by bionic design, named BioLiteNet, is proposed, aimed at enhancing the model’s overall performance in terms of both accuracy and computational efficiency. The model is composed of two key modules: BeeSenseSelector (Channel Attention Screening) and AffScaleConv (Scale-Adaptive Convolutional Fusion). The former mimics the selective attention mechanism observed in honeybee vision for dynamically selecting critical spectral channels, while the latter enables efficient fusion of spatial and spectral features through multi-scale depthwise separable convolution. On multiple hyperspectral benchmark datasets, BioLiteNet is shown to demonstrate outstanding classification performance while maintaining exceptionally low computational costs. Experimental results show that BioLiteNet can maintain high classification accuracy across different datasets, even when using only a small amount of labeled samples. Specifically, it achieves overall accuracies (OA) of 90.02% ± 0.97%, 88.20% ± 5.26%, and 78.64% ± 7.13% on the Indian Pines, Pavia University, and WHU-Hi-LongKou datasets using just 5% of samples, 10% of samples, and 25 samples per class, respectively. Moreover, BioLiteNet consistently requires fewer computational resources than other comparative models. The results indicate that the lightweight hyperspectral image classification model proposed in this study significantly reduces the requirements for computational resources and storage while ensuring classification accuracy, making it well-suited for remote sensing applications under resource constraints. The experimental results further support these findings by demonstrating its robustness and practicality, thereby offering a novel solution for hyperspectral image classification tasks. Full article

► Show Figures

Figure 1

28 pages, 9221 KiB

Open AccessArticle

Adaptive Grid Expected Model Augmentation Based on Golden Section for Maneuvering Extended Object Tracking

by Lifan Sun, Shuo Sun, Dongkai Zhang, Bo Fan and Dan Gao

Remote Sens. 2025, 17(16), 2832; https://doi.org/10.3390/rs17162832 - 14 Aug 2025

Abstract

Maneuvering extended object tracking has garnered significant attention owing to the continuous advancements in the resolution capabilities of modern high-precision radar sensors. The efficacy of tracking algorithms for such objects is heavily contingent upon the design of the model set. However, existing methodologies [...] Read more.

Maneuvering extended object tracking has garnered significant attention owing to the continuous advancements in the resolution capabilities of modern high-precision radar sensors. The efficacy of tracking algorithms for such objects is heavily contingent upon the design of the model set. However, existing methodologies for model set design often yield suboptimal performance when confronted with highly maneuvering extended objects. The expected model augmentation (EMA) algorithm offers a data-driven mechanism for updating the model set in real time. Despite its advantages, the EMA algorithm is constrained by the fixed parameters of its basic models and static transition probabilities between models, thereby limiting its adaptability to extended objects exhibiting complex and dynamic maneuvering behaviors. To address these limitations, this paper proposes a modified variable structure multiple model (VSMM) framework for maneuvering extended object tracking, referred to as the adaptive grid expected model augmentation based on the golden section (GSAG-EMA) algorithm. The approach adaptively adjusts both the model structure and parameters in a grid-based format to accommodate the varying maneuvering patterns. It incorporates both local and global weighting schemes, with two models within the grid based on the golden section. Furthermore, the transition probability matrix is dynamically updated following specific rules, and the execution strategy for each module is determined according to the filtering results. Simulation results under both weak and strong maneuvering scenarios demonstrate that the proposed GSAG-EMA algorithm consistently outperforms the IMM-based, EMA, and AG-BMA algorithms in terms of root mean square error (RMSE) and Hausdorff distance, thereby substantiating its superior tracking performance. Full article

(This article belongs to the Special Issue Radar Data Processing and Analysis)

► Show Figures

Figure 1

22 pages, 9740 KiB

Open AccessArticle

A Novel Error Correction Method for Airborne HRWS SAR Based on Azimuth-Variant Attitude and Range-Variant Doppler Domain Pattern

by Yihao Xu, Fubo Zhang, Longyong Chen, Yangliang Wan and Tao Jiang

Remote Sens. 2025, 17(16), 2831; https://doi.org/10.3390/rs17162831 - 14 Aug 2025

Abstract

In high-resolution and wide-swath (HRWS) synthetic aperture radar (SAR) imaging, the azimuth multi-channel technique effectively suppresses azimuth ambiguity, serving as a reliable approach for achieving wide-swath imaging. However, due to mechanical vibrations of the platform and airflow instabilities, airborne SAR may experience errors [...] Read more.

In high-resolution and wide-swath (HRWS) synthetic aperture radar (SAR) imaging, the azimuth multi-channel technique effectively suppresses azimuth ambiguity, serving as a reliable approach for achieving wide-swath imaging. However, due to mechanical vibrations of the platform and airflow instabilities, airborne SAR may experience errors in attitude and flight path during operation. Furthermore, errors also exist in the antenna patterns, frequency stability, and phase noise among the azimuth multi-channels. The presence of these errors can cause azimuth multi-channel reconstruction failure, resulting in azimuth ambiguity and significantly degrading the quality of HRWS images. This article presents a novel error correction method for airborne HRWS SAR based on azimuth-variant attitude and range-variant Doppler domain pattern, which simultaneously considers the effects of various errors, including channel attitude errors and Doppler domain antenna pattern errors, on azimuth reconstruction. Attitude errors are the primary cause of azimuth-variant errors between channels. This article uses the vector method and attitude transformation matrix to calculate and compensate for the attitude errors of azimuth multi-channels, and employs the two-dimensional frequency-domain echo interferometry method to calculate the fixed delay errors and fixed phase errors. To better achieve channel error compensation, this scheme also considers the estimation and compensation of Doppler domain antenna pattern errors in wide-swath scenes. Finally, the effectiveness of the proposed scheme is confirmed through simulations and processing of airborne real data. Full article

(This article belongs to the Special Issue Advances in Synthetic Aperture Radar Data Processing and Application (Second Edition))

► Show Figures

Figure 1

24 pages, 5458 KiB

Open AccessArticle

Global Prior-Guided Distortion Representation Learning Network for Remote Sensing Image Blind Super-Resolution

by Guanwen Li, Ting Sun, Shijie Yu and Siyao Wu

Remote Sens. 2025, 17(16), 2830; https://doi.org/10.3390/rs17162830 - 14 Aug 2025

Abstract

Most existing deep learning-based super-resolution (SR) methods for remote sensing images rely on predefined degradation assumptions (e.g., bicubic downsampling). However, when real-world degradations deviate from these assumptions, their performance deteriorates significantly. Moreover, explicit degradation estimation approaches based on iterative schemes inevitably lead to [...] Read more.

Most existing deep learning-based super-resolution (SR) methods for remote sensing images rely on predefined degradation assumptions (e.g., bicubic downsampling). However, when real-world degradations deviate from these assumptions, their performance deteriorates significantly. Moreover, explicit degradation estimation approaches based on iterative schemes inevitably lead to accumulated estimation errors and time-consuming processes. In this paper, instead of explicitly estimating degradation types, we first innovatively introduce an MSCN_G coefficient to capture global prior information corresponding to different distortions. Subsequently, distortion-enhanced representations are implicitly estimated through contrastive learning and embedded into a super-resolution network equipped with multiple distortion decoders (D-Decoder). Furthermore, we propose a distortion-related channel segmentation (DCS) strategy that reduces the network’s parameters and computation (FLOPs). We refer to this Global Prior-guided Distortion-enhanced Representation Learning Network as GDRNet. Experiments on both synthetic and real-world remote sensing images demonstrate that our GDRNet outperforms state-of-the-art blind SR methods for remote sensing images in terms of overall performance. Under the experimental condition of anisotropic Gaussian blurring without added noise, with a kernel width of 1.2 and an upscaling factor of 4, the super-resolution reconstruction of remote sensing images on the NWPU-RESISC45 dataset achieves a PSNR of 28.98 dB and SSIM of 0.7656. Full article

(This article belongs to the Special Issue Advanced Applications of Artificial Intelligence in Remote Sensing Image Recognition)

► Show Figures

Figure 1

Journal Description

Remote Sensing

Latest Articles

Journal Menu

Journal Browser

Highly Accessed Articles

Latest Books

E-Mail Alert

News

Topics

Conferences

Special Issues

Topical Collections

Further Information

Guidelines

MDPI Initiatives

Follow MDPI