Previous Issue
Volume 17, July-1
 
 
remotesensing-logo

Journal Browser

Journal Browser

Remote Sens., Volume 17, Issue 14 (July-2 2025) – 24 articles

  • Issues are regarded as officially published after their release is announced to the table of contents alert mailing list.
  • You may sign up for e-mail alerts to receive table of contents of newly released issues.
  • PDF is the official format for papers published in both, html and pdf forms. To view the papers in pdf format, click on the "PDF Full-text" link, and use the free Adobe Reader to open them.
Order results
Result details
Section
Select all
Export citation of selected articles as:
24 pages, 4465 KiB  
Article
A Deep Learning-Based Echo Extrapolation Method by Fusing Radar Mosaic and RMAPS-NOW Data
by Shanhao Wang, Zhiqun Hu, Fuzeng Wang, Ruiting Liu, Lirong Wang and Jiexin Chen
Remote Sens. 2025, 17(14), 2356; https://doi.org/10.3390/rs17142356 (registering DOI) - 9 Jul 2025
Abstract
Radar echo extrapolation is a critical forecasting tool in the field of meteorology, playing an especially vital role in nowcasting and weather modification operations. In recent years, spatiotemporal sequence prediction models based on deep learning have garnered significant attention and achieved notable progress [...] Read more.
Radar echo extrapolation is a critical forecasting tool in the field of meteorology, playing an especially vital role in nowcasting and weather modification operations. In recent years, spatiotemporal sequence prediction models based on deep learning have garnered significant attention and achieved notable progress in radar echo extrapolation. However, most of these extrapolation network architectures are built upon convolutional neural networks, using radar echo images as input. Typically, radar echo intensity values ranging from −5 to 70 dBZ with a resolution of 5 dBZ are converted into 0–255 grayscale images from pseudo-color representations, which inevitably results in the loss of important echo details. Furthermore, as the extrapolation time increases, the smoothing effect inherent to convolution operations leads to increasingly blurred predictions. To address the algorithmic limitations of deep learning-based echo extrapolation models, this study introduces three major improvements: (1) A Deep Convolutional Generative Adversarial Network (DCGAN) is integrated into the ConvLSTM-based extrapolation model to construct a DCGAN-enhanced architecture, significantly improving the quality of radar echo extrapolation; (2) Considering that the evolution of radar echoes is closely related to the surrounding meteorological environment, the study incorporates specific physical variable products from the initial zero-hour field of RMAPS-NOW (the Rapid-update Multiscale Analysis and Prediction System—NOWcasting subsystem), developed by the Institute of Urban Meteorology, China. These variables are encoded jointly with high-resolution (0.5 dB) radar mosaic data to form multiple radar cells as input. A multi-channel radar echo extrapolation network architecture (MR-DCGAN) is then designed based on the DCGAN framework; (3) Since radar echo decay becomes more prominent over longer extrapolation horizons, this study departs from previous approaches that use a single model to extrapolate 120 min. Instead, it customizes time-specific loss functions for spatiotemporal attenuation correction and independently trains 20 separate models to achieve the full 120 min extrapolation. The dataset consists of radar composite reflectivity mosaics over North China within the range of 116.10–117.50°E and 37.77–38.77°N, collected from June to September during 2018–2022. A total of 39,000 data samples were matched with the initial zero-hour fields from RMAPS-NOW, with 80% (31,200 samples) used for training and 20% (7800 samples) for testing. Based on the ConvLSTM and the proposed MR-DCGAN architecture, 20 extrapolation models were trained using four different input encoding strategies. The models were evaluated using the Critical Success Index (CSI), Probability of Detection (POD), and False Alarm Ratio (FAR). Compared to the baseline ConvLSTM-based extrapolation model without physical variables, the models trained with the MR-DCGAN architecture achieved, on average, 18.59%, 8.76%, and 11.28% higher CSI values, 19.46%, 19.21%, and 19.18% higher POD values, and 19.85%, 11.48%, and 9.88% lower FAR values under the 20 dBZ, 30 dBZ, and 35 dBZ reflectivity thresholds, respectively. Among all tested configurations, the model that incorporated three physical variables—relative humidity (rh), u-wind, and v-wind—demonstrated the best overall performance across various thresholds, with CSI and POD values improving by an average of 16.75% and 24.75%, respectively, and FAR reduced by 15.36%. Moreover, the SSIM of the MR-DCGAN models demonstrates a more gradual decline and maintains higher overall values, indicating superior capability in preserving echo structural features. Meanwhile, the comparative experiments demonstrate that the MR-DCGAN (u, v + rh) model outperforms the MR-ConvLSTM (u, v + rh) model in terms of evaluation metrics. In summary, the model trained with the MR-DCGAN architecture effectively enhances the accuracy of radar echo extrapolation. Full article
(This article belongs to the Special Issue Advance of Radar Meteorology and Hydrology II)
Show Figures

Figure 1

20 pages, 3185 KiB  
Article
Radiative Transfer Model-Integrated Approach for Hyperspectral Simulation of Mixed Soil-Vegetation Scenarios and Soil Organic Carbon Estimation
by Asmaa Abdelbaki, Robert Milewski, Mohammadmehdi Saberioon, Katja Berger, José A. M. Demattê and Sabine Chabrillat
Remote Sens. 2025, 17(14), 2355; https://doi.org/10.3390/rs17142355 (registering DOI) - 9 Jul 2025
Abstract
Soils serve as critical carbon reservoirs, playing an essential role in climate change mitigation and agricultural sustainability. Accurate soil property determination relies on soil spectral reflectance data from Earth observation (EO), but current vegetation models often oversimplify soil conditions. This study introduces a [...] Read more.
Soils serve as critical carbon reservoirs, playing an essential role in climate change mitigation and agricultural sustainability. Accurate soil property determination relies on soil spectral reflectance data from Earth observation (EO), but current vegetation models often oversimplify soil conditions. This study introduces a novel approach that combines radiative transfer models (RTMs) with open-access soil spectral libraries to address this challenge. Focusing on conditions of low soil moisture content (SMC), photosynthetic vegetation (PV), and non-photosynthetic vegetation (NPV), the coupled Marmit–Leaf–Canopy (MLC) model is used to simulate early crop growth stages. The MLC model, which integrates MARMIT and PRO4SAIL2, enables the generation of mixed soil–vegetation scenarios. A simulated EO disturbed soil spectral library (DSSL) was created, significantly expanding the EU LUCAS cropland soil spectral library. A 1D convolutional neural network (1D-CNN) was trained on this database to predict Soil Organic Carbon (SOC) content. The results demonstrated relatively high SOC prediction accuracy compared to previous approaches that rely only on RTMs and/or machine learning approaches. Incorporating soil moisture content significantly improved performance over bare soil alone, yielding an R2 of 0.86 and RMSE of 4.05 g/kg, compared to R2 = 0.71 and RMSE = 6.01 g/kg for bare soil. Adding PV slightly reduced accuracy (R2 = 0.71, RMSE = 6.31 g/kg), while the inclusion of NPV alongside moisture led to modest improvement (R2 = 0.74, RMSE = 5.84 g/kg). The most comprehensive model, incorporating bare soil, SMC, PV, and NPV, achieved a balanced performance (R2 = 0.76, RMSE = 5.49 g/kg), highlighting the importance of accounting for all surface components in SOC estimation. While further validation with additional scenarios and SOC prediction methods is needed, these findings demonstrate, for the first time, using radiative-transfer simulations of mixed vegetation-soil-water environments, that an EO-DSSL approach enhances machine learning-based SOC modeling from EO data, improving SOC mapping accuracy. This innovative framework could significantly improve global-scale SOC predictions, supporting the design of next-generation EO products for more accurate carbon monitoring. Full article
Show Figures

Figure 1

27 pages, 7955 KiB  
Article
Land Surface Condition-Driven Emissivity Variation and Its Impact on Diurnal Land Surface Temperature Retrieval Uncertainty
by Lijuan Wang, Ping Yue, Yang Yang, Sha Sha, Die Hu, Xueyuan Ren, Xiaoping Wang, Hui Han and Xiaoyu Jiang
Remote Sens. 2025, 17(14), 2353; https://doi.org/10.3390/rs17142353 (registering DOI) - 9 Jul 2025
Abstract
Land surface emissivity (LSE) is the most critical factor affecting land surface temperature (LST) retrieval. Understanding its variation characteristics is essential, as this knowledge provides fundamental prior constraints for the LST retrieval process. This study utilizes thermal infrared emissivity and hyperspectral data collected [...] Read more.
Land surface emissivity (LSE) is the most critical factor affecting land surface temperature (LST) retrieval. Understanding its variation characteristics is essential, as this knowledge provides fundamental prior constraints for the LST retrieval process. This study utilizes thermal infrared emissivity and hyperspectral data collected from diverse underlying surfaces from 2017 to 2024 to analyze LSE variation characteristics across different surface types, spectral bands, and temporal scales. Key influencing factors are quantified to establish empirical relationships between LSE dynamics and environmental variables. Furthermore, the impact of LSE models on diurnal LST retrieval accuracy is systematically evaluated through comparative experiments, emphasizing the necessity of integrating time-dependent LSE corrections into radiative transfer equations. The results indicate that LSE in the 8–11 µm band is highly sensitive to surface composition, with distinct dual-valley absorption features observed between 8 and 9.5 µm across different soil types, highlighting spectral variability. The 9.6 µm LSE exhibits strong sensitivity to crop growth dynamics, characterized by pronounced absorption valleys linked to vegetation biochemical properties. Beyond soil composition, LSE is significantly influenced by soil moisture, temperature, and vegetation coverage, emphasizing the need for multi-factor parameterization. LSE demonstrates typical diurnal variations, with an amplitude reaching an order of magnitude of 0.01, driven by thermal inertia and environmental interactions. A diurnal LSE retrieval model, integrating time-averaged LSE and diurnal perturbations, was developed based on underlying surface characteristics. This model reduced the root mean square error (RMSE) of LST retrieved from geostationary satellites from 6.02 °C to 2.97 °C, significantly enhancing retrieval accuracy. These findings deepen the understanding of LSE characteristics and provide a scientific basis for refining LST/LSE separation algorithms in thermal infrared remote sensing and for optimizing LSE parameterization schemes in land surface process models for climate and hydrological simulations. Full article
Show Figures

Figure 1

24 pages, 32355 KiB  
Article
Evaluating UAV LiDAR and Field Spectroscopy for Estimating Residual Dry Matter Across Conservation Grazing Lands
by Bruce Markman, H. Scott Butterfield, Janet Franklin, Lloyd Coulter, Moses Katkowski and Daniel Sousa
Remote Sens. 2025, 17(14), 2352; https://doi.org/10.3390/rs17142352 (registering DOI) - 9 Jul 2025
Abstract
Residual dry matter (RDM) is a term used in rangeland management to describe the non-photosynthetic plant material left on the soil surface at the end of the growing season. RDM measurements are used by agencies and conservation entities for managing grazing and fire [...] Read more.
Residual dry matter (RDM) is a term used in rangeland management to describe the non-photosynthetic plant material left on the soil surface at the end of the growing season. RDM measurements are used by agencies and conservation entities for managing grazing and fire fuels. Measuring the RDM using traditional methods is labor-intensive, costly, and subjective, making consistent sampling challenging. Previous studies have assessed the use of multispectral remote sensing to estimate the RDM, but with limited success across space and time. The existing approaches may be improved through the use of spectroscopic (hyperspectral) sensors, capable of capturing the cellulose and lignin present in dry grass, as well as Unmanned Aerial Vehicle (UAV)-mounted Light Detection and Ranging (LiDAR) sensors, capable of capturing centimeter-scale 3D vegetation structures. Here, we evaluate the relationships between the RDM and spectral and LiDAR data across the Jack and Laura Dangermond Preserve (Santa Barbara County, CA, USA), which uses grazing and prescribed fire for rangeland management. The spectral indices did not correlate with the RDM (R2 < 0.1), likely due to complete areal coverage with dense grass. The LiDAR canopy height models performed better for all the samples (R2 = 0.37), with much stronger performance (R2 = 0.81) when using a stratified model to predict the RDM in plots with predominantly standing (as opposed to laying) vegetation. This study demonstrates the potential of UAV LiDAR for direct RDM quantification where vegetation is standing upright, which could help improve RDM mapping and management for rangelands in California and beyond. Full article
Show Figures

Figure 1

21 pages, 4409 KiB  
Article
Differences in Time Comparison and Positioning of BDS-3 PPP-B2b Signal Broadcast Through GEO
by Hongjiao Ma, Jinming Yang, Xiaolong Guan, Jianfeng Wu and Huabing Wu
Remote Sens. 2025, 17(14), 2351; https://doi.org/10.3390/rs17142351 (registering DOI) - 9 Jul 2025
Abstract
The BeiDou-3 Navigation Satellite System (BDS-3) precise point positioning (PPP) service through the B2b signal (PPP-B2b) leverages precise correction data disseminated by satellites to eliminate or mitigate key error sources, including satellite orbit errors, clock biases, and ionospheric delays, thereby enabling high-precision timing [...] Read more.
The BeiDou-3 Navigation Satellite System (BDS-3) precise point positioning (PPP) service through the B2b signal (PPP-B2b) leverages precise correction data disseminated by satellites to eliminate or mitigate key error sources, including satellite orbit errors, clock biases, and ionospheric delays, thereby enabling high-precision timing and positioning. This paper investigates the disparities in time comparison and positioning capabilities associated with the PPP-B2b signals transmitted by the BDS-3 Geostationary Earth Orbit (GEO) satellites (C59 and C61). Three stations in the Asia–Pacific region were selected to establish two time comparison links. The study evaluated the time transfer accuracy of PPP-B2b signals by analyzing orbit and clock corrections from BDS-3 GEO satellites C59 and C61. Using multi-GNSS final products (GBM post-ephemeris) as a reference, the performance of PPP-B2b-based time comparison was assessed. The results indicate that while both satellites achieve comparable time transfer accuracy, C59 demonstrates superior stability and availability compared to C61. Additionally, five stations from the International GNSS Service (IGS) and the International GNSS Monitoring and Assessment System (iGMAS) were selected to assess the positioning accuracy of PPP-B2b corrections transmitted by BDS-3 GEO satellites C59 and C61. Using IGS/iGMAS weekly solution positioning results as a reference, the analysis demonstrates that PPP-B2b enables centimeter-level static positioning and decimeter-level simulated kinematic positioning. Furthermore, C59 achieves higher positioning accuracy than C61. Full article
Show Figures

Figure 1

21 pages, 4829 KiB  
Article
Quantification of MODIS Land Surface Temperature Downscaled by Machine Learning Algorithms
by Qi Su, Xiangchen Meng, Lin Sun and Zhongqiang Guo
Remote Sens. 2025, 17(14), 2350; https://doi.org/10.3390/rs17142350 (registering DOI) - 9 Jul 2025
Abstract
Land Surface Temperature (LST) is essential for understanding the interactions between the land surface and the atmosphere. This study presents a comprehensive evaluation of machine learning (ML)-based downscaling algorithms to enhance the spatial resolution of MODIS LST data from 960 m to 30 [...] Read more.
Land Surface Temperature (LST) is essential for understanding the interactions between the land surface and the atmosphere. This study presents a comprehensive evaluation of machine learning (ML)-based downscaling algorithms to enhance the spatial resolution of MODIS LST data from 960 m to 30 m, leveraging auxiliary variables including vegetation indices, terrain parameters, and land surface reflectance. By establishing non-linear relationships between LST and predictive variables through eXtreme Gradient Boosting (XGBoost) and Random Forest (RF) algorithms, the proposed framework was rigorously validated using in situ measurements across China’s Heihe River Basin. Comparative analyses demonstrated that integrating multiple vegetation indices (e.g., NDVI, SAVI) with terrain factors yielded superior accuracy compared to factors utilizing land surface reflectance or excessive variable combinations. While slope and aspect parameters marginally improved accuracy in mountainous regions, including them degraded performance in flat terrain. Notably, land surface reflectance proved to be ineffective in snow/ice-covered areas, highlighting the need for specialized treatment in cryospheric environments. This work provides a reference for LST downscaling, with significant implications for environmental monitoring and urban heat island investigations. Full article
Show Figures

Figure 1

23 pages, 4200 KiB  
Article
Thermal Multi-Sensor Assessment of the Spatial Sampling Behavior of Urban Landscapes Using 2D Turbulence Indicators
by Gabriel I. Cotlier, Drazen Skokovic, Juan Carlos Jimenez and José Antonio Sobrino
Remote Sens. 2025, 17(14), 2349; https://doi.org/10.3390/rs17142349 (registering DOI) - 9 Jul 2025
Abstract
Understanding spatial variations in land surface temperature (LST) is critical for analyzing urban climate dynamics, especially within the framework of two-dimensional (2D) turbulence theory. This study assesses the spatial sampling behavior of urban thermal fields across eight metropolitan areas, encompassing diverse morphologies, surface [...] Read more.
Understanding spatial variations in land surface temperature (LST) is critical for analyzing urban climate dynamics, especially within the framework of two-dimensional (2D) turbulence theory. This study assesses the spatial sampling behavior of urban thermal fields across eight metropolitan areas, encompassing diverse morphologies, surface materials, and Köppen–Geiger climate zones. We analyzed thermal infrared (TIR) imagery from two remote sensing platforms—MODIS (1 km) and Landsat (30 m)—to evaluate resolution-dependent turbulence indicators such as spectral slopes and breakpoints. Power spectral analysis revealed systematic divergences across spatial scales. Landsat exhibited more negative breakpoint values, indicating a greater ability to capture fine-scale thermal heterogeneity tied to vegetation, buildings, and surface cover. MODIS, in contrast, emphasized broader thermal gradients, suitable for regional-scale assessments. Seasonal differences reinforced the turbulence framework: summer spectra displayed steeper, more variable slopes, reflecting increased thermal activity and surface–atmosphere decoupling. Despite occasional agreement between sensors, spectral metrics remain inherently resolution-dependent. MODIS is better suited for macro-scale thermal structures, while Landsat provides detailed insights into intra-urban processes. Our findings confirm that 2D turbulence indicators are not fully scale-invariant and vary with sensor resolution, season, and urban form. This multi-sensor comparison offers a framework for interpreting LST data in support of climate adaptation, urban design, and remote sensing integration. Full article
Show Figures

Figure 1

23 pages, 9229 KiB  
Article
Magnetopause Boundary Detection Based on a Deep Image Prior Model Using Simulated Lobster-Eye Soft X-Ray Images
by Fei Wei, Zhihui Lyu, Songwu Peng, Rongcong Wang and Tianran Sun
Remote Sens. 2025, 17(14), 2348; https://doi.org/10.3390/rs17142348 (registering DOI) - 9 Jul 2025
Abstract
This study focuses on the problem of identifying and extracting the magnetopause boundary of the Earth’s magnetosphere using the Soft X-ray Imager (SXI) onboard the Solar Wind Magnetosphere Ionosphere Link Explorer (SMILE) mission. The SXI employs lobster-eye optics to perform panoramic imaging of [...] Read more.
This study focuses on the problem of identifying and extracting the magnetopause boundary of the Earth’s magnetosphere using the Soft X-ray Imager (SXI) onboard the Solar Wind Magnetosphere Ionosphere Link Explorer (SMILE) mission. The SXI employs lobster-eye optics to perform panoramic imaging of the magnetosphere based on the Solar Wind Charge Exchange (SWCX) mechanism. However, several factors are expected to hinder future in-orbit observations, including the intrinsically low signal-to-noise ratio (SNR) of soft-X-ray emission, pronounced vignetting, and the non-uniform effective-area distribution of lobster-eye optics. These limitations could severely constrain the accurate interpretation of magnetospheric structures—especially the magnetopause boundary. To address these challenges, a boundary detection approach is developed that combines image calibration with denoising based on deep image prior (DIP). The method begins with calibration procedures to correct for vignetting and effective area variations in the SXI images, thereby restoring the accurate brightness distribution and improving spatial uniformity. Subsequently, a DIP-based denoising technique is introduced, which leverages the structural prior inherent in convolutional neural networks to suppress high-frequency noise without pretraining. This enhances the continuity and recognizability of boundary structures within the image. Experiments use ideal magnetospheric images generated from magnetohydrodynamic (MHD) simulations as reference data. The results demonstrate that the proposed method significantly improves the accuracy of magnetopause boundary identification under medium and high solar wind number density conditions (N = 10–20 cm−3). The extracted boundary curves consistently achieve a normalized mean squared error (NMSE) below 0.05 compared to the reference models. Additionally, the DIP-processed images show notable improvements in peak signal-to-noise ratio (PSNR) and structural similarity index (SSIM), indicating enhanced image quality and structural fidelity. This method provides adequate technical support for the precise extraction of magnetopause boundary structures in soft X-ray observations and holds substantial scientific and practical value. Full article
Show Figures

Figure 1

17 pages, 2928 KiB  
Article
Hybrid Machine Learning Model for Hurricane Power Outage Estimation from Satellite Night Light Data
by Laiyin Zhu and Steven M. Quiring
Remote Sens. 2025, 17(14), 2347; https://doi.org/10.3390/rs17142347 (registering DOI) - 9 Jul 2025
Abstract
Hurricanes can cause massive power outages and pose significant disruptions to society. Accurately monitoring hurricane power outages will improve predictive models and guide disaster emergency management. However, many challenges exist in obtaining high-quality data on hurricane power outages. We systematically evaluated machine learning [...] Read more.
Hurricanes can cause massive power outages and pose significant disruptions to society. Accurately monitoring hurricane power outages will improve predictive models and guide disaster emergency management. However, many challenges exist in obtaining high-quality data on hurricane power outages. We systematically evaluated machine learning (ML) approaches to reconstruct historical hurricane power outages based on high-resolution (1 km) satellite night light observations from the Defense Meteorological Satellite Program (DMSP) and other ancillary information. We found that the two-step hybrid model significantly improved model prediction performance by capturing a substantial portion of the uncertainty in the zero-inflated data. In general, the classification and regression tree-based machine learning models (XGBoost and random forest) demonstrated better performance than the logistic and CNN models in both binary classification and regression models. For example, the xgb+xgb model has 14% less RMSE than the log+cnn model, and the R-squared value is 25 times larger. The Interpretable ML (SHAP value) identified geographic locations, population, and stable and hurricane night light values as important variables in the XGBoost power outage model. These variables also exhibit meaningful physical relationships with power outages. Our study lays the groundwork for monitoring power outages caused by natural disasters using satellite data and machine learning (ML) approaches. Future work should aim to improve the accuracy of power outage estimations and incorporate more hurricanes from the recently available Visible Infrared Imaging Radiometer Suite (VIIRS) Day/Night Band (DNB) data. Full article
Show Figures

Figure 1

27 pages, 7808 KiB  
Article
Phenology-Aware Transformer for Semantic Segmentation of Non-Food Crops from Multi-Source Remote Sensing Time Series
by Xiongwei Guan, Meiling Liu, Shi Cao and Jiale Jiang
Remote Sens. 2025, 17(14), 2346; https://doi.org/10.3390/rs17142346 (registering DOI) - 9 Jul 2025
Abstract
Accurate identification of non-food crops underpins food security by clarifying land-use dynamics, promoting sustainable farming, and guiding efficient resource allocation. Proper identification and management maintain the balance between food and non-food cropping, a prerequisite for ecological sustainability and a healthy agricultural economy. Distinguishing [...] Read more.
Accurate identification of non-food crops underpins food security by clarifying land-use dynamics, promoting sustainable farming, and guiding efficient resource allocation. Proper identification and management maintain the balance between food and non-food cropping, a prerequisite for ecological sustainability and a healthy agricultural economy. Distinguishing large-scale non-food crops—such as oilseed rape, tea, and cotton—remains challenging because their canopy reflectance spectra are similar. This study proposes a novel phenology-aware Vision Transformer Model (PVM) for accurate, large-scale non-food crop classification. PVM incorporates a Phenology-Aware Module (PAM) that fuses multi-source remote-sensing time series with crop-growth calendars. The study area is Hunan Province, China. We collected Sentinel-1 SAR and Sentinel-2 optical imagery (2021–2022) and corresponding ground-truth samples of non-food crops. The model uses a Vision Transformer (ViT) backbone integrated with PAM. PAM dynamically adjusts temporal attention using encoded phenological cues, enabling the network to focus on key growth stages. A parallel Multi-Task Attention Fusion (MTAF) mechanism adaptively combines Sentinel-1 and Sentinel-2 time-series data. The fusion exploits sensor complementarity and mitigates cloud-induced data gaps. The fused spatiotemporal features feed a Transformer-based decoder that performs multi-class semantic segmentation. On the Hunan dataset, PVM achieved an F1-score of 74.84% and an IoU of 61.38%, outperforming MTAF-TST and 2D-U-Net + CLSTM baselines. Cross-regional validation on the Canadian Cropland Dataset confirmed the model’s generalizability, with an F1-score of 71.93% and an IoU of 55.94%. Ablation experiments verified the contribution of each module. Adding PAM raised IoU by 8.3%, whereas including MTAF improved recall by 8.91%. Overall, PVM effectively integrates phenological knowledge with multi-source imagery, delivering accurate and scalable non-food crop classification. Full article
Show Figures

Figure 1

20 pages, 2387 KiB  
Article
Contrastive Learning-Based Hyperspectral Image Target Detection Using a Gated Dual-Path Network
by Jiake Wu, Rong Liu and Nan Wang
Remote Sens. 2025, 17(14), 2345; https://doi.org/10.3390/rs17142345 (registering DOI) - 9 Jul 2025
Abstract
Deep learning-based hyperspectral target detection (HTD) methods often face the challenge of insufficient prior information and difficulty in distinguishing local and global spectral differences. To address these problems, we propose a self-supervised framework that leverages contrastive learning to reduce dependence on prior knowledge, [...] Read more.
Deep learning-based hyperspectral target detection (HTD) methods often face the challenge of insufficient prior information and difficulty in distinguishing local and global spectral differences. To address these problems, we propose a self-supervised framework that leverages contrastive learning to reduce dependence on prior knowledge, called the Gated Dual-Path Network with Contrastive Learning (GDPNCL). In this work, we introduce a novel sample augmentation strategy for deep network training, in which each pixel in the scene is processed using a dual concentric window to generate positive and negative samples. In addition, a Gated Dual-Path Network (GDPN) is proposed to effectively extract and discriminate local and global information from the spectra. Moreover, to mitigate the issue of false negative samples within the same class and to enhance the contrast between negative samples, we design a Weight Information Noise contrastive estimation (WIN) loss. The loss leverages the relationship between samples to further help the model learn representations that effectively distinguish targets from diverse backgrounds. Finally, the trained encoder is subsequently employed to extract features from the prior spectrum and test pixels, and the cosine similarity between them serves as the detection metric. Comprehensive experiments on four challenging hyperspectral datasets demonstrate that the GDPNCL outperforms state-of-the-art methods, highlighting its effectiveness and robustness in HTD. Full article
Show Figures

Figure 1

25 pages, 11278 KiB  
Article
Analysis of Droughts and Floods Evolution and Teleconnection Factors in the Yangtze River Basin Based on GRACE/GFO
by Ruqing Ren, Tatsuya Nemoto, Venkatesh Raghavan, Xianfeng Song and Zheng Duan
Remote Sens. 2025, 17(14), 2344; https://doi.org/10.3390/rs17142344 - 8 Jul 2025
Abstract
In recent years, under the influence of climate change and human activities, droughts and floods have occurred frequently in the Yangtze River Basin (YRB), seriously threatening socioeconomic development and ecological security. The topography and climate of the YRB are complex, so it is [...] Read more.
In recent years, under the influence of climate change and human activities, droughts and floods have occurred frequently in the Yangtze River Basin (YRB), seriously threatening socioeconomic development and ecological security. The topography and climate of the YRB are complex, so it is crucial to develop appropriate drought and flood policies based on the drought and flood characteristics of different sub-basins. This study calculated the water storage deficit index (WSDI) based on the Gravity Recovery and Climate Experiment (GRACE) and GRACE-Follow On (GFO) mascon model, extended WSDI to the bidirectional monitoring of droughts and floods in the YRB, and verified the reliability of WSDI in monitoring hydrological events through historical documented events. Combined with the wavelet method, it revealed the heterogeneity of climate responses in the three sub-basins of the upper, middle, and lower reaches. The results showed the following. (1) Compared and verified with the Standardized Precipitation Evapotranspiration Index (SPEI), self-calibrating Palmer Drought Severity Index (scPDSI), and documented events, WSDI overcame the limitations of traditional indices and had higher reliability. A total of 21 drought events and 18 flood events were identified in the three sub-basins, with the lowest frequency of drought and flood events in the upper reaches. (2) Most areas of the YRB showed different degrees of wetting on the monthly and seasonal scales, and the slowest trend of wetting was in the lower reaches of the YRB. (3) The degree of influence of teleconnection factors in the upper, middle, and lower reaches of the YRB had gradually increased over time, and, in particular, El Niño Southern Oscillation (ENSO) had a significant impact on the droughts and floods. This study provided a new basis for the early warning of droughts and floods in different sub-basins of the YRB. Full article
(This article belongs to the Special Issue Remote Sensing in Natural Resource and Water Environment II)
Show Figures

Figure 1

18 pages, 3618 KiB  
Article
Quality Assessment of Dual-Polarization C-Band SAR Data Influenced by Precipitation Based on Normalized Polarimetric Radar Vegetation Index
by Jisung Geba Chang, Simon Kraatz, Yisok Oh, Feng Gao and Martha Anderson
Remote Sens. 2025, 17(14), 2343; https://doi.org/10.3390/rs17142343 - 8 Jul 2025
Abstract
Advanced Synthetic Aperture Radar (SAR) has become an essential modality in remote sensing, offering all-weather capabilities and sensitivity to vegetation biophysical parameters and surface conditions, while effectively complementing optical sensor data. This study evaluates the impact of precipitation on the Normalized Polarimetric Radar [...] Read more.
Advanced Synthetic Aperture Radar (SAR) has become an essential modality in remote sensing, offering all-weather capabilities and sensitivity to vegetation biophysical parameters and surface conditions, while effectively complementing optical sensor data. This study evaluates the impact of precipitation on the Normalized Polarimetric Radar Vegetation Index (NPRVI) using dual-polarization Sentinel-1 C-band SAR data from agricultural fields at the Beltsville Agricultural Research Center (BARC). Field-measured precipitation and Global Precipitation Measurement (GPM) precipitation datasets were temporally aligned with Sentinel-1 acquisition times to assess the sensitivity of radar signals to precipitation events. NPRVI exhibited a strong sensitivity to precipitation, particularly within the 1 to 7 h prior to the satellite overpass, even for small amounts of precipitation. A quality assessment (QA) framework was developed to flag and correct precipitation-affected radar observations through interpolation. The adjusted NPRVI values, based on the QA framework using precipitation within a 6 h window, showed strong agreement between field- and GPM-derived data, with an RMSE of 0.09 and a relative RMSE of 19.8%, demonstrating that GPM data can serve as a viable alternative for quality adjustment despite its coarse spatial resolution. The adjusted NPRVI for both soybean and corn fields significantly improved the temporal consistency of the time series and closely followed NDVI trends, while also capturing crop-specific seasonal variations, especially during periods of NDVI saturation or limited variability. These findings underscore the value of the proposed radar-based QA framework in enhancing the interpretability of vegetation dynamics. NPRVI, when adjusted for precipitation effects, can serve as a reliable and complementary tool to optical vegetation indices in agricultural and environmental monitoring. Full article
(This article belongs to the Section Remote Sensing in Agriculture and Vegetation)
Show Figures

Figure 1

19 pages, 3553 KiB  
Article
Research on the Autonomous Orbit Determination of Beidou-3 Assisted by Satellite Laser Ranging Technology
by Wei Xiao, Zhengcheng Wu, Zongnan Li, Lei Fan, Shiwei Guo and Yilun Chen
Remote Sens. 2025, 17(14), 2342; https://doi.org/10.3390/rs17142342 - 8 Jul 2025
Abstract
The Beidou Global System (BDS-3) innovatively achieves autonomous navigation using inter-satellite links (ISL) across the entire constellation, but it still faces challenges such as the limitations of the prior constraint orbital accuracy and the overall constellation rotation. The gradual availability of satellite laser [...] Read more.
The Beidou Global System (BDS-3) innovatively achieves autonomous navigation using inter-satellite links (ISL) across the entire constellation, but it still faces challenges such as the limitations of the prior constraint orbital accuracy and the overall constellation rotation. The gradual availability of satellite laser ranging (SLR) data, with advantages of high precision and no ambiguous parameters, can provide new ideas for solving the current problem. This work firstly deduces the mathematical model for orbit determination by combining inter-satellite links and the introduced satellite laser ranging observations, then designs orbit determination experiments with different prior orbit constraints and different observation data, and finally evaluates the impacts of the prior orbits and the introduction of SLR observations from two dimensions: orbit accuracy and constellation rotation. The experimental results using one month of measured data show the following: (1) There is good consistency among different days, and the accuracy of the prior orbits affects the performance of the orbit determination and the consistency. Compared with broadcast ephemerides, using precise ephemerides as prior constraints significantly improves the consistency, and the orbit accuracy can be increased by about 75%. (2) The type of observation data affects the performance of the orbit determination. Introducing SLR observations can improve the orbit accuracy by approximately 13% to 26%. (3) Regardless of whether broadcast ephemerides or precise ephemerides are used as prior constraints, the constellation translation and rotation still exist after introducing SLR observations. Among the translation parameters, TX is the largest, followed by TY, and TZ is the smallest; all three rotation parameters (RX, RY, and RZ) show relatively large values, which may be related to the limited number of available satellite laser ranging stations during this period. (4) After considering the constellation translation and rotation, the orbit accuracy under different prior constraints remains at the same level. The statistical root mean square error (RMSE) indicates that the orbit accuracy of inclined geosynchronous orbit (IGSO) satellites in three directions is better than 20 cm, while the accuracy of medium earth orbit (MEO) satellites in along-track, cross-track, and radial directions is better than 10 cm, 8 cm, and 5 cm, respectively. Full article
Show Figures

Figure 1

20 pages, 6538 KiB  
Article
A Deep Reinforcement Learning Method with a Low Intercept Probability in a Netted Synthetic Aperture Radar
by Longhao Xie, Ziyang Cheng, Ming Li and Huiyong Li
Remote Sens. 2025, 17(14), 2341; https://doi.org/10.3390/rs17142341 - 8 Jul 2025
Abstract
A deep reinforcement learning (DRL)-based power allocation method is proposed to achieve a low probability of intercept (LPI) in a netted synthetic aperture radar (SAR). To provide a physically meaningful and intuitive assessment of a netted radar for LPI performance, a netted circular [...] Read more.
A deep reinforcement learning (DRL)-based power allocation method is proposed to achieve a low probability of intercept (LPI) in a netted synthetic aperture radar (SAR). To provide a physically meaningful and intuitive assessment of a netted radar for LPI performance, a netted circular equivalent vulnerable radius (NCEVR) is proposed and adopted. For SAR detection performance, the resolution, signal-to-noise ratio in a single pulse, and signal-to-noise ratio in SAR imaging are integrated at the task level. The LPI performance is achieved by minimizing NCEVR within the constraints of SAR detection performance. The powers in multiple moments are optimized using the DRL proximal policy optimization algorithm with the designed reward and observation. A DRL-based solver is provided for LPI radar, which handles problems that are difficult to optimize using traditional methods. The effectiveness is verified by simulations. Full article
(This article belongs to the Special Issue Array and Signal Processing for Radar (Second Edition))
Show Figures

Figure 1

31 pages, 6764 KiB  
Article
Upscaling Frameworks Drive Prediction Accuracy and Uncertainty When Mapping Aboveground Biomass Density from the Synergism of Spaceborne LiDAR, SAR, and Passive Optical Data
by Inacio T. Bueno, Carlos A. Silva, Monique B. Schlickmann, Victoria M. Donovan, Jeff W. Atkins, Kody M. Brock, Jinyi Xia, Denis R. Valle, Jiangxiao Qiu, Jason Vogel, Andres Susaeta, Ajay Sharma, Carine Klauberg, Midhun Mohan and Ana Paula Dalla Corte
Remote Sens. 2025, 17(14), 2340; https://doi.org/10.3390/rs17142340 - 8 Jul 2025
Viewed by 46
Abstract
Accurate mapping of aboveground biomass density (AGBD) is vital for ecological research and carbon cycle monitoring. Integrating multi-source remote sensing data offers significant potential to enhance the accuracy and coverage of AGBD estimates. This study evaluated three upscaling frameworks for integrating GEDI LiDAR, [...] Read more.
Accurate mapping of aboveground biomass density (AGBD) is vital for ecological research and carbon cycle monitoring. Integrating multi-source remote sensing data offers significant potential to enhance the accuracy and coverage of AGBD estimates. This study evaluated three upscaling frameworks for integrating GEDI LiDAR, SAR, and optical satellite data to create wall-to-wall AGBD maps. The frameworks tested in this paper were: (1) a single-step approach using optical imagery, (2) a two-stage approach with GEDI-derived variables, and (3) a three-stage approach combining imagery and in situ-derived allometries. Internal validation showed that framework 1 achieved the lowest root mean square difference (%RMSD) of 53.3% and highest coefficient of determination (R2) of 0.53. An independent external validation of the AGBD map was performed using in situ observations, also revealing that framework 1 was the most accurate (%RMSD = 39.3% and R2 = 0.93), while frameworks 2 and 3 were less accurate (%RMSD = 54.7, 44.7 and R2 = 0.95, 0.90, respectively). Herein, we show that upscaling frameworks significantly impacted AGBD map uncertainty and the magnitude of estimate differences. Our findings suggest that upscaling framework 1 based on a single step approach was the most effective for capturing detailed AGBD variations, while careful consideration of model sensitivity and map uncertainties is essential for reliable AGBD estimation. This study provides valuable insights for advancing forest AGBD monitoring and highlights the potential for further enhancements in remote sensing methodologies. Full article
(This article belongs to the Section Forest Remote Sensing)
Show Figures

Figure 1

22 pages, 6546 KiB  
Article
Remote Sensing-Based Assessment of Evapotranspiration Patterns in a UNESCO World Heritage Site Under Increasing Water Competition
by Maria C. Moyano, Monica Garcia, Luis Juana, Laura Recuero, Lucia Tornos, Joshua B. Fisher, Néstor Fernández and Alicia Palacios-Orueta
Remote Sens. 2025, 17(14), 2339; https://doi.org/10.3390/rs17142339 - 8 Jul 2025
Viewed by 3
Abstract
In water-scarce regions, natural ecosystems and agriculture increasingly compete for limited water resources, intensifying stress during periods of drought. To assess these competing demands, we applied a modified PT-JPL model that incorporates the thermal inertial approach as a substitute for relative humidity ( [...] Read more.
In water-scarce regions, natural ecosystems and agriculture increasingly compete for limited water resources, intensifying stress during periods of drought. To assess these competing demands, we applied a modified PT-JPL model that incorporates the thermal inertial approach as a substitute for relative humidity (RH) in estimating soil evaporation—a method that significantly outperforms the original PT-JPL formulation in Mediterranean semi-arid irrigated areas. This remote sensing framework enabled us to quantify spatial and temporal variations in water use across both natural and agricultural systems within the UNESCO World Heritage site of Doñana. Our analysis revealed an increasing evapotranspiration (ET) trend in intensified agricultural areas and rice fields surrounding the National Park (R = 0.3), contrasted by a strong negative ET trend in wetlands (R < −0.5). These opposing patterns suggest a growing diversion of water toward irrigation at the expense of natural ecosystems. The impact was especially marked during droughts, such as the 2011–2016 period, when precipitation declined by 16%. In wetlands, ET was significantly correlated with precipitation (R > 0.4), highlighting their vulnerability to reduced water inputs. These findings offer crucial insights to support sustainable water management strategies that balance agricultural productivity with the preservation of ecologically valuable systems under mounting climatic and anthropogenic pressures typical of semi-arid Mediterranean environments. Full article
Show Figures

Figure 1

46 pages, 5911 KiB  
Article
Leveraging Prior Knowledge in Semi-Supervised Learning for Precise Target Recognition
by Guohao Xie, Zhe Chen, Yaan Li, Mingsong Chen, Feng Chen, Yuxin Zhang, Hongyan Jiang and Hongbing Qiu
Remote Sens. 2025, 17(14), 2338; https://doi.org/10.3390/rs17142338 - 8 Jul 2025
Viewed by 35
Abstract
Underwater acoustic target recognition (UATR) is challenged by complex marine noise, scarce labeled data, and inadequate multi-scale feature extraction in conventional methods. This study proposes DART-MT, a semi-supervised framework that integrates a Dual Attention Parallel Residual Network Transformer with a mean teacher paradigm, [...] Read more.
Underwater acoustic target recognition (UATR) is challenged by complex marine noise, scarce labeled data, and inadequate multi-scale feature extraction in conventional methods. This study proposes DART-MT, a semi-supervised framework that integrates a Dual Attention Parallel Residual Network Transformer with a mean teacher paradigm, enhanced by domain-specific prior knowledge. The architecture employs a Convolutional Block Attention Module (CBAM) for localized feature refinement, a lightweight New Transformer Encoder for global context modeling, and a novel TriFusion Block to synergize spectral–temporal–spatial features through parallel multi-branch fusion, addressing the limitations of single-modality extraction. Leveraging the mean teacher framework, DART-MT optimizes consistency regularization to exploit unlabeled data, effectively mitigating class imbalance and annotation scarcity. Evaluations on the DeepShip and ShipsEar datasets demonstrate state-of-the-art accuracy: with 10% labeled data, DART-MT achieves 96.20% (DeepShip) and 94.86% (ShipsEar), surpassing baseline models by 7.2–9.8% in low-data regimes, while reaching 98.80% (DeepShip) and 98.85% (ShipsEar) with 90% labeled data. Under varying noise conditions (−20 dB to 20 dB), the model maintained a robust performance (F1-score: 92.4–97.1%) with 40% lower variance than its competitors, and ablation studies validated each module’s contribution (TriFusion Block alone improved accuracy by 6.9%). This research advances UATR by (1) resolving multi-scale feature fusion bottlenecks, (2) demonstrating the efficacy of semi-supervised learning in marine acoustics, and (3) providing an open-source implementation for reproducibility. In future work, we will extend cross-domain adaptation to diverse oceanic environments. Full article
Show Figures

Figure 1

25 pages, 8372 KiB  
Article
CSDNet: Context-Aware Segmentation of Disaster Aerial Imagery Using Detection-Guided Features and Lightweight Transformers
by Ahcene Zetout and Mohand Saïd Allili
Remote Sens. 2025, 17(14), 2337; https://doi.org/10.3390/rs17142337 - 8 Jul 2025
Viewed by 34
Abstract
Accurate multi-class semantic segmentation of disaster-affected areas is essential for rapid response and effective recovery planning. We present CSDNet, a context-aware segmentation model tailored to disaster scene scenarios, designed to improve segmentation of both large-scale disaster zones and small, underrepresented classes. The architecture [...] Read more.
Accurate multi-class semantic segmentation of disaster-affected areas is essential for rapid response and effective recovery planning. We present CSDNet, a context-aware segmentation model tailored to disaster scene scenarios, designed to improve segmentation of both large-scale disaster zones and small, underrepresented classes. The architecture combines a lightweight transformer module for global context modeling with depthwise separable convolutions (DWSCs) to enhance efficiency without compromising representational capacity. Additionally, we introduce a detection-guided feature fusion mechanism that integrates outputs from auxiliary detection tasks to mitigate class imbalance and improve discrimination of visually similar categories. Extensive experiments on several public datasets demonstrate that our model significantly improves segmentation of both man-made infrastructure and natural damage-related features, offering a robust and efficient solution for post-disaster analysis. Full article
Show Figures

Figure 1

21 pages, 3305 KiB  
Article
Unlocking Potato Phenology: Harnessing Sentinel-1 and Sentinel-2 Synergy for Precise Crop Stage Detection
by Diego Gomez, Pablo Salvador, Jorge Gil and Juan Fernando Rodrigo
Remote Sens. 2025, 17(14), 2336; https://doi.org/10.3390/rs17142336 - 8 Jul 2025
Viewed by 43
Abstract
Global challenges such as climate change and population growth require improvements in crop monitoring models. To address these issues, this study advances the identification of potato crop phenological stages using satellite remote sensing, a field where cereals have been the primary focus. We [...] Read more.
Global challenges such as climate change and population growth require improvements in crop monitoring models. To address these issues, this study advances the identification of potato crop phenological stages using satellite remote sensing, a field where cereals have been the primary focus. We introduce a methodology using Sentinel-1 (S1) and Sentinel-2 (S2) time series data to pinpoint critical phenological stages—emergence, canopy closure, flowering, senescence onset, and harvest timing—at the field scale. Our approach utilizes analysis of NDVI, fAPAR, and IRECI2 from S2, alongside VH and VV polarizations from S1, informed by domain knowledge of the spectral and morphological responses of potato crops. We propose the integration of NDVI and VH indices, NDVI_VH, to improve stage detection accuracy. Comparative analysis with ground-observed stages validated the method’s effectiveness, with NDVI proving to be one of the most informative indices, achieving RMSEs of 12 and 14 days for emergence and closure, and 17 days for the onset of senescence. The integrated NDVI_VH approach complemented NDVI, particularly in harvest and flowering stages, where VH enhanced accuracy, achieving an overall R2 value of 0.80. The study demonstrates the potential of combining SAR and optical data for post-season crop phenology analysis, providing insights that can inform the development of new methods and strategies to enhance on-season crop monitoring and yield forecasting. Full article
(This article belongs to the Special Issue Remote Sensing for Precision Farming and Crop Phenology)
Show Figures

Figure 1

18 pages, 4682 KiB  
Article
UAS Remote Sensing for Coastal Wetland Vegetation Biomass Estimation: A Destructive vs. Non-Destructive Sampling Experiment
by Grayson R. Morgan, Lane Stevenson, Cuizhen Wang and Ram Avtar
Remote Sens. 2025, 17(14), 2335; https://doi.org/10.3390/rs17142335 - 8 Jul 2025
Viewed by 45
Abstract
Coastal wetlands are critical ecosystems that require effective monitoring to support conservation and restoration efforts. This study evaluates the use of small unmanned aerial systems (sUAS) and multispectral imagery to estimate aboveground biomass (AGB) in tidal marshes, comparing models calibrated with destructive versus [...] Read more.
Coastal wetlands are critical ecosystems that require effective monitoring to support conservation and restoration efforts. This study evaluates the use of small unmanned aerial systems (sUAS) and multispectral imagery to estimate aboveground biomass (AGB) in tidal marshes, comparing models calibrated with destructive versus non-destructive in situ sampling methods. Imagery was collected over South Carolina’s North Inlet-Winyah Bay National Estuarine Research Reserve, and vegetation indices (VIs) were derived from sUAS imagery to model biomass. Stepwise linear regression was used to develop and validate models based on both sampling approaches. Destructive sampling models, particularly those using the Normalized Difference Vegetation Index (NDVI) and Difference Vegetation Index (DVI), achieved the lowest root mean square error (RMSE) values (as low as 70.91 g/m2), indicating higher predictive accuracy. Non-destructive models, while less accurate (minimum RMSE of 214.86 g/m2), demonstrated higher R2 values (0.44 and 0.61), suggesting the potential for broader application with further refinement. These findings highlight the trade-offs between ecological impact and model performance, and support the viability of non-destructive methods for biomass estimation in sensitive wetland environments. Future work should explore machine learning approaches and improved temporal alignment of data collection to enhance model robustness. Full article
Show Figures

Figure 1

20 pages, 3609 KiB  
Article
Beyond the Grid: GLRT-Based TomoSAR Fast Detection for Retrieving Height and Thermal Dilation
by Nabil Haddad, Karima Hadj-Rabah, Alessandra Budillon and Gilda Schirinzi
Remote Sens. 2025, 17(14), 2334; https://doi.org/10.3390/rs17142334 - 8 Jul 2025
Viewed by 69
Abstract
The Tomographic Synthetic Aperture Radar (TomoSAR) technique is widely used for monitoring urban infrastructures, as it enables the mapping of individual scatterers across additional dimensions such as height (3D), thermal dilation (4D), and deformation velocity (5D). Retrieving this information is crucial for building [...] Read more.
The Tomographic Synthetic Aperture Radar (TomoSAR) technique is widely used for monitoring urban infrastructures, as it enables the mapping of individual scatterers across additional dimensions such as height (3D), thermal dilation (4D), and deformation velocity (5D). Retrieving this information is crucial for building management and maintenance. Nevertheless, accurately extracting it from TomoSAR data poses several challenges, particularly the presence of outliers due to uneven and limited baseline distributions. One way to address these issues is through statistical detection approaches such as the Generalized Likelihood Ratio Test, which ensures a Constant False Alarm Rate. While effective, these methods face two primary limitations: high computational complexity and the off-grid problem caused by the discretization of the search space. To overcome these drawbacks, we propose an approach that combines a quick initialization process using Fast-Sup GLRT with local descent optimization. This method operates directly in the continuous domain, bypassing the limitations of grid-based search while significantly reducing computational costs. Experiments conducted on both simulated and real datasets acquired with the TerraSAR-X satellite over the Spanish city of Barcelona demonstrate the ability of the proposed approach to maintain computational efficiency while improving scatterer localization accuracy in the third and fourth dimensions. Full article
(This article belongs to the Section Urban Remote Sensing)
Show Figures

Graphical abstract

24 pages, 5310 KiB  
Article
Deep Learning-Driven Multi-Temporal Detection: Leveraging DeeplabV3+/Efficientnet-B08 Semantic Segmentation for Deforestation and Forest Fire Detection
by Joe Soundararajan, Andrew Kalukin, Jordan Malof and Dong Xu
Remote Sens. 2025, 17(14), 2333; https://doi.org/10.3390/rs17142333 - 8 Jul 2025
Viewed by 65
Abstract
Deforestation and forest fires are escalating global threats that require timely, scalable, and cost-effective monitoring systems. While UAV and ground-based solutions offer fine-grained data, they are often constrained by limited spatial coverage, high operational costs, and logistical challenges. In contrast, satellite imagery provides [...] Read more.
Deforestation and forest fires are escalating global threats that require timely, scalable, and cost-effective monitoring systems. While UAV and ground-based solutions offer fine-grained data, they are often constrained by limited spatial coverage, high operational costs, and logistical challenges. In contrast, satellite imagery provides broad, repeatable, and economically feasible coverage. This study presents a deep learning framework that combines the DeepLabV3+ architecture with an EfficientNet-B08 backbone to address both deforestation and wildfire detection using satellite imagery. The system utilizes advanced multi-scale feature extraction and Group Normalization to enable robust semantic segmentation under challenging atmospheric conditions and complex forest structures. It is evaluated on two benchmark datasets. In the Amazon forest segmentation dataset, the model achieves a validation Intersection over Union (IoU) of 0.9100 and a pixel accuracy of 0.9605, demonstrating strong performance in delineating forest boundaries. In FireDataset_20m, which presents a severe class imbalance between fire and non-fire pixels, the framework achieves 99.95% accuracy, 93.16% precision, and 91.47% recall. A qualitative analysis confirms the model’s ability to accurately localize fire hotspots and deforested areas. These results highlight the model’s dual-purpose utility for high-resolution, multi-temporal environmental monitoring. Its balanced performance across metrics and adaptability to complex terrain conditions make it a promising tool for supporting forest conservation, early fire detection, and evidence-based policy interventions. Full article
Show Figures

Figure 1

24 pages, 2201 KiB  
Article
TIMA-Net: A Lightweight Remote Sensing Image Change Detection Network Based on Temporal Interaction Enhancement and Multi-Scale Aggregation
by Zhijun Zhou, Xuejie Zhang, Xiaoliang Luo, Lvchun Wang, Wei Yu, Shufang Xu and Longbao Wang
Remote Sens. 2025, 17(14), 2332; https://doi.org/10.3390/rs17142332 - 8 Jul 2025
Viewed by 35
Abstract
Remote sensing image change detection (RSCD) holds significant application value in fields such as environmental monitoring, post-disaster assessment, and urban planning. However, existing deep learning methods often face challenges of high computational complexity and insufficient detail capture, particularly demonstrating limited performance in detecting [...] Read more.
Remote sensing image change detection (RSCD) holds significant application value in fields such as environmental monitoring, post-disaster assessment, and urban planning. However, existing deep learning methods often face challenges of high computational complexity and insufficient detail capture, particularly demonstrating limited performance in detecting high-resolution images and complex change regions. To address these issues, this paper proposes a novel network architecture, TIMA-Net, which is designed for efficient remote sensing image change detection. By introducing the timing interaction enhancement module and the progressive decoder based on multi-scale fusion, TIMA-Net improves the accuracy of change detection while ensuring efficient computing performance. Specifically, TIMA-Net designs a temporal interaction enhancement module based on dual-branch and coordinate attention and combines channel segmentation and multi-scale features to enhance the representation ability of the changed region. The comparative experimental results show that TIMA-Net outperforms several state-of-the-art methods on multiple remote sensing datasets, especially in terms of accuracy and computational efficiency. The ablation experiment results show that each module contributes to the final performance. In summary, TIMA-Net not only provides an efficient and accurate remote sensing image change detection solution but also shows its strong potential and broad application prospects in practical applications. Full article
Show Figures

Graphical abstract

Previous Issue
Back to TopTop