Accurate and efficient detection of building changes in remote sensing imagery is crucial for urban planning,disaster emergency response,and resource management.However,existing methods face challenges such as spectra...Accurate and efficient detection of building changes in remote sensing imagery is crucial for urban planning,disaster emergency response,and resource management.However,existing methods face challenges such as spectral similarity between buildings and backgrounds,sensor variations,and insufficient computational efficiency.To address these challenges,this paper proposes a novel Multi-scale Efficient Wavelet-based Change Detection Network(MewCDNet),which integrates the advantages of Convolutional Neural Networks and Transformers,balances computational costs,and achieves high-performance building change detection.The network employs EfficientNet-B4 as the backbone for hierarchical feature extraction,integrates multi-level feature maps through a multi-scale fusion strategy,and incorporates two key modules:Cross-temporal Difference Detection(CTDD)and Cross-scale Wavelet Refinement(CSWR).CTDD adopts a dual-branch architecture that combines pixel-wise differencing with semanticaware Euclidean distance weighting to enhance the distinction between true changes and background noise.CSWR integrates Haar-based Discrete Wavelet Transform with multi-head cross-attention mechanisms,enabling cross-scale feature fusion while significantly improving edge localization and suppressing spurious changes.Extensive experiments on four benchmark datasets demonstrate MewCDNet’s superiority over comparison methods:achieving F1 scores of 91.54%on LEVIR,93.70%on WHUCD,and 64.96%on S2Looking for building change detection.Furthermore,MewCDNet exhibits optimal performance on the multi-class⋅SYSU dataset(F1:82.71%),highlighting its exceptional generalization capability.展开更多
Deep learning has made significant progress in the field of oriented object detection for remote sensing images.However,existing methods still face challenges when dealing with difficult tasks such as multi-scale targ...Deep learning has made significant progress in the field of oriented object detection for remote sensing images.However,existing methods still face challenges when dealing with difficult tasks such as multi-scale targets,complex backgrounds,and small objects in remote sensing.Maintaining model lightweight to address resource constraints in remote sensing scenarios while improving task completion for remote sensing tasks remains a research hotspot.Therefore,we propose an enhanced multi-scale feature extraction lightweight network EM-YOLO based on the YOLOv8s architecture,specifically optimized for the characteristics of large target scale variations,diverse orientations,and numerous small objects in remote sensing images.Our innovations lie in two main aspects:First,a dynamic snake convolution(DSC)is introduced into the backbone network to enhance the model’s feature extraction capability for oriented targets.Second,an innovative focusing-diffusion module is designed in the feature fusion neck to effectively integrate multi-scale feature information.Finally,we introduce Layer-Adaptive Sparsity for magnitude-based Pruning(LASP)method to perform lightweight network pruning to better complete tasks in resource-constrained scenarios.Experimental results on the lightweight platform Orin demonstrate that the proposed method significantly outperforms the original YOLOv8s model in oriented remote sensing object detection tasks,and achieves comparable or superior performance to state-of-the-art methods on three authoritative remote sensing datasets(DOTA v1.0,DOTA v1.5,and HRSC2016).展开更多
An improved model based on you only look once version 8(YOLOv8)is proposed to solve the problem of low detection accuracy due to the diversity of object sizes in optical remote sensing images.Firstly,the feature pyram...An improved model based on you only look once version 8(YOLOv8)is proposed to solve the problem of low detection accuracy due to the diversity of object sizes in optical remote sensing images.Firstly,the feature pyramid network(FPN)structure of the original YOLOv8 mode is replaced by the generalized-FPN(GFPN)structure in GiraffeDet to realize the"cross-layer"and"cross-scale"adaptive feature fusion,to enrich the semantic information and spatial information on the feature map to improve the target detection ability of the model.Secondly,a pyramid-pool module of multi atrous spatial pyramid pooling(MASPP)is designed by using the idea of atrous convolution and feature pyramid structure to extract multi-scale features,so as to improve the processing ability of the model for multi-scale objects.The experimental results show that the detection accuracy of the improved YOLOv8 model on DIOR dataset is 92%and mean average precision(mAP)is 87.9%,respectively 3.5%and 1.7%higher than those of the original model.It is proved the detection and classification ability of the proposed model on multi-dimensional optical remote sensing target has been improved.展开更多
As a vital food crop,rice is an important part of global food crops.Studying the spatiotemporal changes in rice cultivation facilitates early prediction of production risks and provides support for agricultural policy...As a vital food crop,rice is an important part of global food crops.Studying the spatiotemporal changes in rice cultivation facilitates early prediction of production risks and provides support for agricultural policy decisions related to rice.With the increasing application of satellite remote sensing technology in crop monitoring,remote sensing for rice cultivation has emerged as a novel approach,offering new perspectives for monitoring rice planting.This paper briefly outlined the current research and development status of satellite remote sensing for monitoring rice cultivation both at home and abroad.Foreign scholars have made innovations in data sources and methodologies for satellite remote sensing monitoring,and utilized multi-source satellite information and machine learning algorithms to enhance the accuracy of rice planting monitoring.Scholars in China have achieved significant results in the study of satellite remote sensing for monitoring rice cultivation.Their research and application in monitoring rice planting areas provide valuable references for agricultural production management.However,satellite remote sensing monitoring of rice still faces challenges such as low spatiotemporal resolution and difficulties related to cloud cover and data fusion,which require further in-depth investigation.Additionally,there are shortcomings in the accuracy of remote sensing monitoring for fragmented farmland plots and smallholder farming.To address these issues,future efforts should focus on developing multi-source heterogeneous data fusion analysis technologies and researching monitoring systems.These advancements are expected to enable high-precision large-scale acquisition of rice planting information,laying a foundation for future smart agriculture.展开更多
High-resolution remote sensing images(HRSIs)are now an essential data source for gathering surface information due to advancements in remote sensing data capture technologies.However,their significant scale changes an...High-resolution remote sensing images(HRSIs)are now an essential data source for gathering surface information due to advancements in remote sensing data capture technologies.However,their significant scale changes and wealth of spatial details pose challenges for semantic segmentation.While convolutional neural networks(CNNs)excel at capturing local features,they are limited in modeling long-range dependencies.Conversely,transformers utilize multihead self-attention to integrate global context effectively,but this approach often incurs a high computational cost.This paper proposes a global-local multiscale context network(GLMCNet)to extract both global and local multiscale contextual information from HRSIs.A detail-enhanced filtering module(DEFM)is proposed at the end of the encoder to refine the encoder outputs further,thereby enhancing the key details extracted by the encoder and effectively suppressing redundant information.In addition,a global-local multiscale transformer block(GLMTB)is proposed in the decoding stage to enable the modeling of rich multiscale global and local information.We also design a stair fusion mechanism to transmit deep semantic information from deep to shallow layers progressively.Finally,we propose the semantic awareness enhancement module(SAEM),which further enhances the representation of multiscale semantic features through spatial attention and covariance channel attention.Extensive ablation analyses and comparative experiments were conducted to evaluate the performance of the proposed method.Specifically,our method achieved a mean Intersection over Union(mIoU)of 86.89%on the ISPRS Potsdam dataset and 84.34%on the ISPRS Vaihingen dataset,outperforming existing models such as ABCNet and BANet.展开更多
Desert shrubs are indispensable in maintaining ecological stability by reducing soil erosion,enhancing water retention,and boosting soil fertility,which are critical factors in mitigating desertification processes.Due...Desert shrubs are indispensable in maintaining ecological stability by reducing soil erosion,enhancing water retention,and boosting soil fertility,which are critical factors in mitigating desertification processes.Due to the complex topography,variable climate,and challenges in field surveys in desert regions,this paper proposes YOLO-Desert-Shrub(YOLO-DS),a detection method for identifying desert shrubs in UAV remote sensing images based on an enhanced YOLOv8n framework.This method accurately identifying shrub species,locations,and coverage.To address the issue of small individual plants dominating the dataset,the SPDconv convolution module is introduced in the Backbone and Neck layers of the YOLOv8n model,replacing conventional convolutions.This structural optimization mitigates information degradation in fine-grained data while strengthening discriminative feature capture across spatial scales within desert shrub datasets.Furthermore,a structured state-space model is integrated into the main network,and the MambaLayer is designed to dynamically extract and refine shrub-specific features from remote sensing images,effectively filtering out background noise and irrelevant interference to enhance feature representation.Benchmark evaluations reveal the YOLO-DS framework attains 79.56%mAP40weight,demonstrating 2.2%absolute gain versus the baseline YOLOv8n architecture,with statistically significant advantages over contemporary detectors in cross-validation trials.The predicted plant coverage exhibits strong consistency with manually measured coverage,with a coefficient of determination(R^(2))of 0.9148 and a Root Mean Square Error(RMSE)of1.8266%.The proposed UAV-based remote sensing method utilizing the YOLO-DS effectively identify and locate desert shrubs,monitor canopy sizes and distribution,and provide technical support for automated desert shrub monitoring.展开更多
Agricultural greenhouses(AGHs)are increasingly used globally to control the crop growth environment,which are vital for food production,resource conservation,and rural economies.Advances in high-quality data acquisiti...Agricultural greenhouses(AGHs)are increasingly used globally to control the crop growth environment,which are vital for food production,resource conservation,and rural economies.Advances in high-quality data acquisition methods and information retrieval algorithms have improved the ability to extract AGHs from remote sensing images(e.g.,satellite and uncrewed aerial vehicle(UAV)).Research on this topic began in 1989,and the number of related studies has increased annually.This paper provides a review of the development of remote sensing of AGHs and research hotspots.It summarizes the current status and trends of data sources,identification features,methods,and accuracy of AGHs extraction.Due to the unique spectral,textural,and geometric characteristics of AGHs,research studies have primarily utilized optical remote sensing data from sensors with spatial resolutions of 30 m or more,such as Landsat,Sentinel,Gaofen(GF),and Worldview,to extract AGHs.Machine learning and deep learning methods have provided more precise results for extracting AGHs than threshold segmentation methods.In contrast,deep learning algorithms have been primarily used with high-spatial resolution data and small-scale study areas,with accuracy rates generally exceeding 90.00%.However,future research may use higher spatial resolution images to improve the accuracy and detail of AGH extraction.Recent studies have integrated multiple data sources and performed time-series analysis to improve monitoring of dynamic changes in AGHs.Moreover,emphasis should be placed on optimizing data fusion techniques,implementing sample transfer methods,expanding the number of sensors,and increasing the application of artificial intelligence(AI)in monitoring AGHs.These efforts will provide more reliable methods and tools to improve agricultural production and resource utilization efficiency.This review provides resources for researchers and decision-makers involved in modern agricultural development,as well as scientific evidence for the sustainable development of rural areas.展开更多
Remote sensing image super-resolution technology is pivotal for enhancing image quality in critical applications including environmental monitoring,urban planning,and disaster assessment.However,traditional methods ex...Remote sensing image super-resolution technology is pivotal for enhancing image quality in critical applications including environmental monitoring,urban planning,and disaster assessment.However,traditional methods exhibit deficiencies in detail recovery and noise suppression,particularly when processing complex landscapes(e.g.,forests,farmlands),leading to artifacts and spectral distortions that limit practical utility.To address this,we propose an enhanced Super-Resolution Generative Adversarial Network(SRGAN)framework featuring three key innovations:(1)Replacement of L1/L2 loss with a robust Charbonnier loss to suppress noise while preserving edge details via adaptive gradient balancing;(2)A multi-loss joint optimization strategy dynamically weighting Charbonnier loss(β=0.5),Visual Geometry Group(VGG)perceptual loss(α=1),and adversarial loss(γ=0.1)to synergize pixel-level accuracy and perceptual quality;(3)A multi-scale residual network(MSRN)capturing cross-scale texture features(e.g.,forest canopies,mountain contours).Validated on Sentinel-2(10 m)and SPOT-6/7(2.5 m)datasets covering 904 km2 in Motuo County,Xizang,our method outperforms the SRGAN baseline(SR4RS)with Peak Signal-to-Noise Ratio(PSNR)gains of 0.29 dB and Structural Similarity Index(SSIM)improvements of 3.08%on forest imagery.Visual comparisons confirm enhanced texture continuity despite marginal Learned Perceptual Image Patch Similarity(LPIPS)increases.The method significantly improves noise robustness and edge retention in complex geomorphology,demonstrating 18%faster response in forest fire early warning and providing high-resolution support for agricultural/urban monitoring.Future work will integrate spectral constraints and lightweight architectures.展开更多
An improved Carnegie Ames Stanford Approach （CASA） model based on two kinds of remote sensing （RS） data, Landsat Enhanced Thematic Mapper Plus （ETM ＋） and Moderate Resolution Imaging Spectro- radiometer （MODIS...An improved Carnegie Ames Stanford Approach （CASA） model based on two kinds of remote sensing （RS） data, Landsat Enhanced Thematic Mapper Plus （ETM ＋） and Moderate Resolution Imaging Spectro- radiometer （MODIS）, and climate variables were applied to estimate the Net Primary Productivity （NPP） of Xuzhou in June of each year from 2001 to 2010. The NPP of the study area decreased as the spatial scale increased. The average NPP of terrestrial vegetation in Xuzhou showed a decreasing trend in recent years, likely due to changes in climate and environment. The study area was divided into four sub-regions, designated as highest, moderately high, moderately low, and lowest in NPR The area designated as the lowest sub-region in NPP increased with expanding scale, indicating that the NPP distribution varied with different spatial scales. The NPP of different vegetation types was also significantly influenced by scale. In particular, the NPP of urban woodland produced lower estimates because of mixed pixels. Similar trends in NPP were observed with different RS data. In addition, expansion of residential areas and reduction of vegetated areas were the major reasons for NPP change. Land cover changes in urban areas reduced NPP, which could chiefly be attributed to human-induced disturbance.展开更多
One of the study objectives of global change is land use/cover change (LUCC) by using multiscale remotely sensed data on global and regional scale. In this paper, field sample, digital camera, Landsat-ETM+ (ETM+, Enha...One of the study objectives of global change is land use/cover change (LUCC) by using multiscale remotely sensed data on global and regional scale. In this paper, field sample, digital camera, Landsat-ETM+ (ETM+, Enhanced Thematic Mapper) image and the National Oceanic and Atmospheric Administration/the advanced very high resolution radiometer (NOAA/AVHRR) image were integrated to detect, simulate and analyze the vegetation fractional coverage of typical steppe in northern China. The results show: (1) Vegetation fractional coverage measured by digital camera is more precise than results measured by other methods. It can be used to validate other measuring results. (2) Vegetation fractional coverage measured by 1 m 2 field sample change fluctuantly for different observers and for different sample areas. In this experiment, the coverage is generally high compared with the result measured by digital camera, and the average absolute error is 9.92%, but two groups measure results, correlation coefficient r(2) = 0.89. (3) Three kinds of methods using remotely sensed data were adopted to simulate the vegetation fractional coverage. Average absolute errors of the vegetation fractional coverage, measured by ETM+ and NOAA, are respectively 7.03% and 7.83% compared with the result measured by digital camera. When NOAA pixel was decomposed by ETM+ pixels after geometrical registry, the average absolute errors measured by this method is 5.68% compared with the digital camera result. Correction coefficients of three results with digital camera result r(2) are respectively 0.78, 0.61 and 0.76. (4) The result of statistic model established by NOAA-NDVI (NDVI, Normalized Difference Vegetation Index) and the vegetation fractional coverage measured by digital camera show lower precision (r(2) = 0.65) than the result of statistic model established by ETM+-NDVI and digital camera coverage then converted to NOAA image (r(2) = 0.80). Pixel decomposability method improves the precision of measuring the vegetation fractional coverage on a large scale. This is a significant practice on scaling by using remotely sensed data. Integrated application of multi-scale remotely sensed data in earth observation will be an important approach to promoting measuring precision of ecological parameters.展开更多
Although the Convolutional Neural Network(CNN)has shown great potential for land cover classification,the frequently used single-scale convolution kernel limits the scope of informa-tion extraction.Therefore,we propos...Although the Convolutional Neural Network(CNN)has shown great potential for land cover classification,the frequently used single-scale convolution kernel limits the scope of informa-tion extraction.Therefore,we propose a Multi-Scale Fully Convolutional Network(MSFCN)with a multi-scale convolutional kernel as well as a Channel Attention Block(CAB)and a Global Pooling Module(GPM)in this paper to exploit discriminative representations from two-dimensional(2D)satellite images.Meanwhile,to explore the ability of the proposed MSFCN for spatio-temporal images,we expand our MSFCN to three-dimension using three-dimensional(3D)CNN,capable of harnessing each land cover category’s time series interac-tion from the reshaped spatio-temporal remote sensing images.To verify the effectiveness of the proposed MSFCN,we conduct experiments on two spatial datasets and two spatio-temporal datasets.The proposed MSFCN achieves 60.366%on the WHDLD dataset and 75.127%on the GID dataset in terms of mIoU index while the figures for two spatio-temporal datasets are 87.753%and 77.156%.Extensive comparative experiments and abla-tion studies demonstrate the effectiveness of the proposed MSFCN.展开更多
This paper aims at providing multi-source remote sensing images registered in geometric space for image fusion.Focusing on the characteristics and differences of multi-source remote sensing images,a feature-based regi...This paper aims at providing multi-source remote sensing images registered in geometric space for image fusion.Focusing on the characteristics and differences of multi-source remote sensing images,a feature-based registration algorithm is implemented.The key technologies include image scale-space for implementing multi-scale properties,Harris corner detection for keypoints extraction,and partial intensity invariant feature descriptor(PIIFD)for keypoints description.Eventually,a multi-scale Harris-PIIFD image registration algorithm framework is proposed.The experimental results of fifteen sets of representative real data show that the algorithm has excellent,stable performance in multi-source remote sensing image registration,and can achieve accurate spatial alignment,which has strong practical application value and certain generalization ability.展开更多
The degradation of optical remote sensing images due to atmospheric haze poses a significant obstacle,profoundly impeding their effective utilization across various domains.Dehazing methodologies have emerged as pivot...The degradation of optical remote sensing images due to atmospheric haze poses a significant obstacle,profoundly impeding their effective utilization across various domains.Dehazing methodologies have emerged as pivotal components of image preprocessing,fostering an improvement in the quality of remote sensing imagery.This enhancement renders remote sensing data more indispensable,thereby enhancing the accuracy of target iden-tification.Conventional defogging techniques based on simplistic atmospheric degradation models have proven inadequate for mitigating non-uniform haze within remotely sensed images.In response to this challenge,a novel UNet Residual Attention Network(URA-Net)is proposed.This paradigmatic approach materializes as an end-to-end convolutional neural network distinguished by its utilization of multi-scale dense feature fusion clusters and gated jump connections.The essence of our methodology lies in local feature fusion within dense residual clusters,enabling the extraction of pertinent features from both preceding and current local data,depending on contextual demands.The intelligently orchestrated gated structures facilitate the propagation of these features to the decoder,resulting in superior outcomes in haze removal.Empirical validation through a plethora of experiments substantiates the efficacy of URA-Net,demonstrating its superior performance compared to existing methods when applied to established datasets for remote sensing image defogging.On the RICE-1 dataset,URA-Net achieves a Peak Signal-to-Noise Ratio(PSNR)of 29.07 dB,surpassing the Dark Channel Prior(DCP)by 11.17 dB,the All-in-One Network for Dehazing(AOD)by 7.82 dB,the Optimal Transmission Map and Adaptive Atmospheric Light For Dehazing(OTM-AAL)by 5.37 dB,the Unsupervised Single Image Dehazing(USID)by 8.0 dB,and the Superpixel-based Remote Sensing Image Dehazing(SRD)by 8.5 dB.Particularly noteworthy,on the SateHaze1k dataset,URA-Net attains preeminence in overall performance,yielding defogged images characterized by consistent visual quality.This underscores the contribution of the research to the advancement of remote sensing technology,providing a robust and efficient solution for alleviating the adverse effects of haze on image quality.展开更多
INTRODUCTION.On May 1st,2024,around 2:10 a.m.,a catastrophic collapse occurred along the Meilong Expressway near Meizhou City,Guangdong Province,China,at coordinates 24°29′24″N and 116°40′25″E.This colla...INTRODUCTION.On May 1st,2024,around 2:10 a.m.,a catastrophic collapse occurred along the Meilong Expressway near Meizhou City,Guangdong Province,China,at coordinates 24°29′24″N and 116°40′25″E.This collapse resulted in a pavement failure of approximately 17.9 m in length and covering an area of about 184.3 m^(2)(Chinanews,2024).展开更多
With the increasing global population and mounting pressures on agricultural production,precise pest monitoring has become a critical factor in ensuring food security.Traditional monitoring methods,often inefficient,s...With the increasing global population and mounting pressures on agricultural production,precise pest monitoring has become a critical factor in ensuring food security.Traditional monitoring methods,often inefficient,struggle to meet the demands of modern agriculture.Drone remote sensing technology,leveraging its high efficiency and flexibility,demonstrates significant potential in pest monitoring.Equipped with multispectral,hyperspectral,and thermal infrared sensors,drones can rapidly cover large agricultural fields,capturing high-resolution imagery and data to detect spectral variations in crops.This enables effective differentiation between healthy and infested plants,facilitating early pest identification and targeted control.This paper systematically reviews the current applications of drone remote sensing technology in pest monitoring by examining different sensor types and their use in monitoring major crop pests and diseases.It also discusses existing challenges,aiming to provide insights and references for future research.展开更多
Accurate fine-grained geospatial scene classification using remote sensing imagery is essential for a wide range of applications.However,existing approaches often rely on manually zooming remote sensing images at diff...Accurate fine-grained geospatial scene classification using remote sensing imagery is essential for a wide range of applications.However,existing approaches often rely on manually zooming remote sensing images at different scales to create typical scene samples.This approach fails to adequately support the fixed-resolution image interpretation requirements in real-world scenarios.To address this limitation,we introduce the million-scale fine-grained geospatial scene classification dataset(MEET),which contains over 1.03 million zoom-free remote sensing scene samples,manually annotated into 80 fine-grained categories.In MEET,each scene sample follows a scene-in-scene layout,where the central scene serves as the reference,and auxiliary scenes provide crucial spatial context for fine-grained classification.Moreover,to tackle the emerging challenge of scene-in-scene classification,we present the context-aware transformer(CAT),a model specifically designed for this task,which adaptively fuses spatial context to accurately classify the scene samples.CAT adaptively fuses spatial context to accurately classify the scene samples by learning attentional features that capture the relationships between the center and auxiliary scenes.Based on MEET,we establish a comprehensive benchmark for fine-grained geospatial scene classification,evaluating CAT against 11 competitive baselines.The results demonstrate that CAT significantly outperforms these baselines,achieving a 1.88%higher balanced accuracy(BA)with the Swin-Large backbone,and a notable 7.87%improvement with the Swin-Huge backbone.Further experiments validate the effectiveness of each module in CAT and show the practical applicability of CAT in the urban functional zone mapping.The source code and dataset will be publicly available at https://jerrywyn.github.io/project/MEET.html.展开更多
Yellow rust(Puccinia striiformis f.sp.Tritici,YR)and fusarium head blight(Fusarium graminearum,FHB)are the two main diseases affecting wheat in the main grain-producing areas of East China,which is common for the two ...Yellow rust(Puccinia striiformis f.sp.Tritici,YR)and fusarium head blight(Fusarium graminearum,FHB)are the two main diseases affecting wheat in the main grain-producing areas of East China,which is common for the two diseases to appear simultaneously in some main production areas.It is necessary to discriminate wheat YR and FHB at the regional scale to accurately locate the disease in space,conduct detailed disease severity monitoring,and scientific control.Four images on different dates were acquired from Sentinel-2,Landsat-8,and Gaofen-1 during the critical period of winter wheat,and 22 remote sensing features that characterize the wheat growth status were then calculated.Meanwhile,6 meteorological parameters that reflect the wheat phenological information were also obtained by combining the site meteorological data and spatial interpolation technology.Then,the principal components(PCs)of comprehensive remote sensing and meteorological features were extracted with principal component analysis(PCA).The PCs-based discrimination models were established to map YR and FHB damage using the random forest(RF)and backpropagation neural network(BPNN).The models’performance was verified based on the disease field truth data(57 plots during the filling period)and 5-fold cross-validation.The results revealed that the PCs obtained after PCA dimensionality reduction outperformed the initial features(IFs)from remote sensing and meteorology in discriminating between the two diseases.Compared to the IFs,the average area under the curve for both micro-average and macro-average ROC curves increased by 0.07 in the PCs-based RF models and increased by 0.16 and 0.13,respectively,in the PCs-based BPNN models.Notably,the PCs-based BPNN discrimination model emerged as the most effective,achieving an overall accuracy of 83.9%.Our proposed discrimination model for wheat YR and FHB,coupled with multi-source remote sensing images and meteorological data,overcomes the limitations of a single-sensor and single-phase remote sensing information in multiple stress discrimination in cloudy and rainy areas.It performs well in revealing the damage spatial distribution of the two diseases at a regional scale,providing a basis for detailed disease severity monitoring,and scientific prevention and control.展开更多
Accurate estimation of understory terrain has significant scientific importance for maintaining ecosystem balance and biodiversity conservation.Addressing the issue of inadequate representation of spatial heterogeneit...Accurate estimation of understory terrain has significant scientific importance for maintaining ecosystem balance and biodiversity conservation.Addressing the issue of inadequate representation of spatial heterogeneity when traditional forest topographic inversion methods consider the entire forest as the inversion unit,this study pro⁃poses a differentiated modeling approach to forest types based on refined land cover classification.Taking Puerto Ri⁃co and Maryland as study areas,a multi-dimensional feature system is constructed by integrating multi-source re⁃mote sensing data:ICESat-2 spaceborne LiDAR is used to obtain benchmark values for understory terrain,topo⁃graphic factors such as slope and aspect are extracted based on SRTM data,and vegetation cover characteristics are analyzed using Landsat-8 multispectral imagery.This study incorporates forest type as a classification modeling con⁃dition and applies the random forest algorithm to build differentiated topographic inversion models.Experimental re⁃sults indicate that,compared to traditional whole-area modeling methods(RMSE=5.06 m),forest type-based classi⁃fication modeling significantly improves the accuracy of understory terrain estimation(RMSE=2.94 m),validating the effectiveness of spatial heterogeneity modeling.Further sensitivity analysis reveals that canopy structure parame⁃ters(with RMSE variation reaching 4.11 m)exert a stronger regulatory effect on estimation accuracy compared to forest cover,providing important theoretical support for optimizing remote sensing models of forest topography.展开更多
Cloud detection is a critical preprocessing step in remote sensing image processing, as the presence of clouds significantly affects the accuracy of remote sensing data and limits its applicability across various doma...Cloud detection is a critical preprocessing step in remote sensing image processing, as the presence of clouds significantly affects the accuracy of remote sensing data and limits its applicability across various domains. This study presents an enhanced cloud detection method based on the U-Net architecture, designed to address the challenges of multi-scale cloud features and long-range dependencies inherent in remote sensing imagery. A Multi-Scale Dilated Attention (MSDA) module is introduced to effectively integrate multi-scale information and model long-range dependencies across different scales, enhancing the model’s ability to detect clouds of varying sizes. Additionally, a Multi-Head Self-Attention (MHSA) mechanism is incorporated to improve the model’s capacity for capturing finer details, particularly in distinguishing thin clouds from surface features. A multi-path supervision mechanism is also devised to ensure the model learns cloud features at multiple scales, further boosting the accuracy and robustness of cloud mask generation. Experimental results demonstrate that the enhanced model achieves superior performance compared to other benchmarked methods in complex scenarios. It significantly improves cloud detection accuracy, highlighting its strong potential for practical applications in cloud detection tasks.展开更多
In recent years,convolutional neural networks(CNN)and Transformer architectures have made significant progress in the field of remote sensing(RS)change detection(CD).Most of the existing methods directly stack multipl...In recent years,convolutional neural networks(CNN)and Transformer architectures have made significant progress in the field of remote sensing(RS)change detection(CD).Most of the existing methods directly stack multiple layers of Transformer blocks,which achieves considerable improvement in capturing variations,but at a rather high computational cost.We propose a channel-Efficient Change Detection Network(CE-CDNet)to address the problems of high computational cost and imbalanced detection accuracy in remote sensing building change detection.The adaptive multi-scale feature fusion module(CAMSF)and lightweight Transformer decoder(LTD)are introduced to improve the change detection effect.The CAMSF module can adaptively fuse multi-scale features to improve the model’s ability to detect building changes in complex scenes.In addition,the LTD module reduces computational costs and maintains high detection accuracy through an optimized self-attention mechanism and dimensionality reduction operation.Experimental test results on three commonly used remote sensing building change detection data sets show that CE-CDNet can reduce a certain amount of computational overhead while maintaining detection accuracy comparable to existing mainstream models,showing good performance advantages.展开更多
基金supported by the Henan Province Key R&D Project under Grant 241111210400the Henan Provincial Science and Technology Research Project under Grants 252102211047,252102211062,252102211055 and 232102210069+2 种基金the Jiangsu Provincial Scheme Double Initiative Plan JSS-CBS20230474,the XJTLU RDF-21-02-008the Science and Technology Innovation Project of Zhengzhou University of Light Industry under Grant 23XNKJTD0205the Higher Education Teaching Reform Research and Practice Project of Henan Province under Grant 2024SJGLX0126。
文摘Accurate and efficient detection of building changes in remote sensing imagery is crucial for urban planning,disaster emergency response,and resource management.However,existing methods face challenges such as spectral similarity between buildings and backgrounds,sensor variations,and insufficient computational efficiency.To address these challenges,this paper proposes a novel Multi-scale Efficient Wavelet-based Change Detection Network(MewCDNet),which integrates the advantages of Convolutional Neural Networks and Transformers,balances computational costs,and achieves high-performance building change detection.The network employs EfficientNet-B4 as the backbone for hierarchical feature extraction,integrates multi-level feature maps through a multi-scale fusion strategy,and incorporates two key modules:Cross-temporal Difference Detection(CTDD)and Cross-scale Wavelet Refinement(CSWR).CTDD adopts a dual-branch architecture that combines pixel-wise differencing with semanticaware Euclidean distance weighting to enhance the distinction between true changes and background noise.CSWR integrates Haar-based Discrete Wavelet Transform with multi-head cross-attention mechanisms,enabling cross-scale feature fusion while significantly improving edge localization and suppressing spurious changes.Extensive experiments on four benchmark datasets demonstrate MewCDNet’s superiority over comparison methods:achieving F1 scores of 91.54%on LEVIR,93.70%on WHUCD,and 64.96%on S2Looking for building change detection.Furthermore,MewCDNet exhibits optimal performance on the multi-class⋅SYSU dataset(F1:82.71%),highlighting its exceptional generalization capability.
基金funded by the Hainan Province Science and Technology Special Fund under Grant ZDYF2024GXJS292.
文摘Deep learning has made significant progress in the field of oriented object detection for remote sensing images.However,existing methods still face challenges when dealing with difficult tasks such as multi-scale targets,complex backgrounds,and small objects in remote sensing.Maintaining model lightweight to address resource constraints in remote sensing scenarios while improving task completion for remote sensing tasks remains a research hotspot.Therefore,we propose an enhanced multi-scale feature extraction lightweight network EM-YOLO based on the YOLOv8s architecture,specifically optimized for the characteristics of large target scale variations,diverse orientations,and numerous small objects in remote sensing images.Our innovations lie in two main aspects:First,a dynamic snake convolution(DSC)is introduced into the backbone network to enhance the model’s feature extraction capability for oriented targets.Second,an innovative focusing-diffusion module is designed in the feature fusion neck to effectively integrate multi-scale feature information.Finally,we introduce Layer-Adaptive Sparsity for magnitude-based Pruning(LASP)method to perform lightweight network pruning to better complete tasks in resource-constrained scenarios.Experimental results on the lightweight platform Orin demonstrate that the proposed method significantly outperforms the original YOLOv8s model in oriented remote sensing object detection tasks,and achieves comparable or superior performance to state-of-the-art methods on three authoritative remote sensing datasets(DOTA v1.0,DOTA v1.5,and HRSC2016).
基金supported by the National Natural Science Foundation of China(No.62241109)the Tianjin Science and Technology Commissioner Project(No.20YDTPJC01110)。
文摘An improved model based on you only look once version 8(YOLOv8)is proposed to solve the problem of low detection accuracy due to the diversity of object sizes in optical remote sensing images.Firstly,the feature pyramid network(FPN)structure of the original YOLOv8 mode is replaced by the generalized-FPN(GFPN)structure in GiraffeDet to realize the"cross-layer"and"cross-scale"adaptive feature fusion,to enrich the semantic information and spatial information on the feature map to improve the target detection ability of the model.Secondly,a pyramid-pool module of multi atrous spatial pyramid pooling(MASPP)is designed by using the idea of atrous convolution and feature pyramid structure to extract multi-scale features,so as to improve the processing ability of the model for multi-scale objects.The experimental results show that the detection accuracy of the improved YOLOv8 model on DIOR dataset is 92%and mean average precision(mAP)is 87.9%,respectively 3.5%and 1.7%higher than those of the original model.It is proved the detection and classification ability of the proposed model on multi-dimensional optical remote sensing target has been improved.
基金Supported by Natural Science Foundation General Project of Heilongjiang Province(C2018050).
文摘As a vital food crop,rice is an important part of global food crops.Studying the spatiotemporal changes in rice cultivation facilitates early prediction of production risks and provides support for agricultural policy decisions related to rice.With the increasing application of satellite remote sensing technology in crop monitoring,remote sensing for rice cultivation has emerged as a novel approach,offering new perspectives for monitoring rice planting.This paper briefly outlined the current research and development status of satellite remote sensing for monitoring rice cultivation both at home and abroad.Foreign scholars have made innovations in data sources and methodologies for satellite remote sensing monitoring,and utilized multi-source satellite information and machine learning algorithms to enhance the accuracy of rice planting monitoring.Scholars in China have achieved significant results in the study of satellite remote sensing for monitoring rice cultivation.Their research and application in monitoring rice planting areas provide valuable references for agricultural production management.However,satellite remote sensing monitoring of rice still faces challenges such as low spatiotemporal resolution and difficulties related to cloud cover and data fusion,which require further in-depth investigation.Additionally,there are shortcomings in the accuracy of remote sensing monitoring for fragmented farmland plots and smallholder farming.To address these issues,future efforts should focus on developing multi-source heterogeneous data fusion analysis technologies and researching monitoring systems.These advancements are expected to enable high-precision large-scale acquisition of rice planting information,laying a foundation for future smart agriculture.
基金provided by the Science Research Project of Hebei Education Department under grant No.BJK2024115.
文摘High-resolution remote sensing images(HRSIs)are now an essential data source for gathering surface information due to advancements in remote sensing data capture technologies.However,their significant scale changes and wealth of spatial details pose challenges for semantic segmentation.While convolutional neural networks(CNNs)excel at capturing local features,they are limited in modeling long-range dependencies.Conversely,transformers utilize multihead self-attention to integrate global context effectively,but this approach often incurs a high computational cost.This paper proposes a global-local multiscale context network(GLMCNet)to extract both global and local multiscale contextual information from HRSIs.A detail-enhanced filtering module(DEFM)is proposed at the end of the encoder to refine the encoder outputs further,thereby enhancing the key details extracted by the encoder and effectively suppressing redundant information.In addition,a global-local multiscale transformer block(GLMTB)is proposed in the decoding stage to enable the modeling of rich multiscale global and local information.We also design a stair fusion mechanism to transmit deep semantic information from deep to shallow layers progressively.Finally,we propose the semantic awareness enhancement module(SAEM),which further enhances the representation of multiscale semantic features through spatial attention and covariance channel attention.Extensive ablation analyses and comparative experiments were conducted to evaluate the performance of the proposed method.Specifically,our method achieved a mean Intersection over Union(mIoU)of 86.89%on the ISPRS Potsdam dataset and 84.34%on the ISPRS Vaihingen dataset,outperforming existing models such as ABCNet and BANet.
基金supported by the National Public Welfare Forest Desert Shrubbery Monitoring Project。
文摘Desert shrubs are indispensable in maintaining ecological stability by reducing soil erosion,enhancing water retention,and boosting soil fertility,which are critical factors in mitigating desertification processes.Due to the complex topography,variable climate,and challenges in field surveys in desert regions,this paper proposes YOLO-Desert-Shrub(YOLO-DS),a detection method for identifying desert shrubs in UAV remote sensing images based on an enhanced YOLOv8n framework.This method accurately identifying shrub species,locations,and coverage.To address the issue of small individual plants dominating the dataset,the SPDconv convolution module is introduced in the Backbone and Neck layers of the YOLOv8n model,replacing conventional convolutions.This structural optimization mitigates information degradation in fine-grained data while strengthening discriminative feature capture across spatial scales within desert shrub datasets.Furthermore,a structured state-space model is integrated into the main network,and the MambaLayer is designed to dynamically extract and refine shrub-specific features from remote sensing images,effectively filtering out background noise and irrelevant interference to enhance feature representation.Benchmark evaluations reveal the YOLO-DS framework attains 79.56%mAP40weight,demonstrating 2.2%absolute gain versus the baseline YOLOv8n architecture,with statistically significant advantages over contemporary detectors in cross-validation trials.The predicted plant coverage exhibits strong consistency with manually measured coverage,with a coefficient of determination(R^(2))of 0.9148 and a Root Mean Square Error(RMSE)of1.8266%.The proposed UAV-based remote sensing method utilizing the YOLO-DS effectively identify and locate desert shrubs,monitor canopy sizes and distribution,and provide technical support for automated desert shrub monitoring.
基金Under the auspices of the Strategic Priority Research Program of the Chinese Academy of Sciences(No.XDA28050400)Jilin Province Key Research and Development Project(No.20230202040NC)Common Application Support Platform for National Civil Space Infrastructure Land Observation Satellites(No.2017-000052-73-01-001735)。
文摘Agricultural greenhouses(AGHs)are increasingly used globally to control the crop growth environment,which are vital for food production,resource conservation,and rural economies.Advances in high-quality data acquisition methods and information retrieval algorithms have improved the ability to extract AGHs from remote sensing images(e.g.,satellite and uncrewed aerial vehicle(UAV)).Research on this topic began in 1989,and the number of related studies has increased annually.This paper provides a review of the development of remote sensing of AGHs and research hotspots.It summarizes the current status and trends of data sources,identification features,methods,and accuracy of AGHs extraction.Due to the unique spectral,textural,and geometric characteristics of AGHs,research studies have primarily utilized optical remote sensing data from sensors with spatial resolutions of 30 m or more,such as Landsat,Sentinel,Gaofen(GF),and Worldview,to extract AGHs.Machine learning and deep learning methods have provided more precise results for extracting AGHs than threshold segmentation methods.In contrast,deep learning algorithms have been primarily used with high-spatial resolution data and small-scale study areas,with accuracy rates generally exceeding 90.00%.However,future research may use higher spatial resolution images to improve the accuracy and detail of AGH extraction.Recent studies have integrated multiple data sources and performed time-series analysis to improve monitoring of dynamic changes in AGHs.Moreover,emphasis should be placed on optimizing data fusion techniques,implementing sample transfer methods,expanding the number of sensors,and increasing the application of artificial intelligence(AI)in monitoring AGHs.These efforts will provide more reliable methods and tools to improve agricultural production and resource utilization efficiency.This review provides resources for researchers and decision-makers involved in modern agricultural development,as well as scientific evidence for the sustainable development of rural areas.
基金This study was supported by:Inner Mongolia Academy of Forestry Sciences Open Research Project(Grant No.KF2024MS03)The Project to Improve the Scientific Research Capacity of the Inner Mongolia Academy of Forestry Sciences(Grant No.2024NLTS04)The Innovation and Entrepreneurship Training Program for Undergraduates of Beijing Forestry University(Grant No.X202410022268).
文摘Remote sensing image super-resolution technology is pivotal for enhancing image quality in critical applications including environmental monitoring,urban planning,and disaster assessment.However,traditional methods exhibit deficiencies in detail recovery and noise suppression,particularly when processing complex landscapes(e.g.,forests,farmlands),leading to artifacts and spectral distortions that limit practical utility.To address this,we propose an enhanced Super-Resolution Generative Adversarial Network(SRGAN)framework featuring three key innovations:(1)Replacement of L1/L2 loss with a robust Charbonnier loss to suppress noise while preserving edge details via adaptive gradient balancing;(2)A multi-loss joint optimization strategy dynamically weighting Charbonnier loss(β=0.5),Visual Geometry Group(VGG)perceptual loss(α=1),and adversarial loss(γ=0.1)to synergize pixel-level accuracy and perceptual quality;(3)A multi-scale residual network(MSRN)capturing cross-scale texture features(e.g.,forest canopies,mountain contours).Validated on Sentinel-2(10 m)and SPOT-6/7(2.5 m)datasets covering 904 km2 in Motuo County,Xizang,our method outperforms the SRGAN baseline(SR4RS)with Peak Signal-to-Noise Ratio(PSNR)gains of 0.29 dB and Structural Similarity Index(SSIM)improvements of 3.08%on forest imagery.Visual comparisons confirm enhanced texture continuity despite marginal Learned Perceptual Image Patch Similarity(LPIPS)increases.The method significantly improves noise robustness and edge retention in complex geomorphology,demonstrating 18%faster response in forest fire early warning and providing high-resolution support for agricultural/urban monitoring.Future work will integrate spectral constraints and lightweight architectures.
文摘An improved Carnegie Ames Stanford Approach （CASA） model based on two kinds of remote sensing （RS） data, Landsat Enhanced Thematic Mapper Plus （ETM ＋） and Moderate Resolution Imaging Spectro- radiometer （MODIS）, and climate variables were applied to estimate the Net Primary Productivity （NPP） of Xuzhou in June of each year from 2001 to 2010. The NPP of the study area decreased as the spatial scale increased. The average NPP of terrestrial vegetation in Xuzhou showed a decreasing trend in recent years, likely due to changes in climate and environment. The study area was divided into four sub-regions, designated as highest, moderately high, moderately low, and lowest in NPR The area designated as the lowest sub-region in NPP increased with expanding scale, indicating that the NPP distribution varied with different spatial scales. The NPP of different vegetation types was also significantly influenced by scale. In particular, the NPP of urban woodland produced lower estimates because of mixed pixels. Similar trends in NPP were observed with different RS data. In addition, expansion of residential areas and reduction of vegetated areas were the major reasons for NPP change. Land cover changes in urban areas reduced NPP, which could chiefly be attributed to human-induced disturbance.
文摘One of the study objectives of global change is land use/cover change (LUCC) by using multiscale remotely sensed data on global and regional scale. In this paper, field sample, digital camera, Landsat-ETM+ (ETM+, Enhanced Thematic Mapper) image and the National Oceanic and Atmospheric Administration/the advanced very high resolution radiometer (NOAA/AVHRR) image were integrated to detect, simulate and analyze the vegetation fractional coverage of typical steppe in northern China. The results show: (1) Vegetation fractional coverage measured by digital camera is more precise than results measured by other methods. It can be used to validate other measuring results. (2) Vegetation fractional coverage measured by 1 m 2 field sample change fluctuantly for different observers and for different sample areas. In this experiment, the coverage is generally high compared with the result measured by digital camera, and the average absolute error is 9.92%, but two groups measure results, correlation coefficient r(2) = 0.89. (3) Three kinds of methods using remotely sensed data were adopted to simulate the vegetation fractional coverage. Average absolute errors of the vegetation fractional coverage, measured by ETM+ and NOAA, are respectively 7.03% and 7.83% compared with the result measured by digital camera. When NOAA pixel was decomposed by ETM+ pixels after geometrical registry, the average absolute errors measured by this method is 5.68% compared with the digital camera result. Correction coefficients of three results with digital camera result r(2) are respectively 0.78, 0.61 and 0.76. (4) The result of statistic model established by NOAA-NDVI (NDVI, Normalized Difference Vegetation Index) and the vegetation fractional coverage measured by digital camera show lower precision (r(2) = 0.65) than the result of statistic model established by ETM+-NDVI and digital camera coverage then converted to NOAA image (r(2) = 0.80). Pixel decomposability method improves the precision of measuring the vegetation fractional coverage on a large scale. This is a significant practice on scaling by using remotely sensed data. Integrated application of multi-scale remotely sensed data in earth observation will be an important approach to promoting measuring precision of ecological parameters.
基金supported by the National Natural Science Foundation of China[grant number 41671452].
文摘Although the Convolutional Neural Network(CNN)has shown great potential for land cover classification,the frequently used single-scale convolution kernel limits the scope of informa-tion extraction.Therefore,we propose a Multi-Scale Fully Convolutional Network(MSFCN)with a multi-scale convolutional kernel as well as a Channel Attention Block(CAB)and a Global Pooling Module(GPM)in this paper to exploit discriminative representations from two-dimensional(2D)satellite images.Meanwhile,to explore the ability of the proposed MSFCN for spatio-temporal images,we expand our MSFCN to three-dimension using three-dimensional(3D)CNN,capable of harnessing each land cover category’s time series interac-tion from the reshaped spatio-temporal remote sensing images.To verify the effectiveness of the proposed MSFCN,we conduct experiments on two spatial datasets and two spatio-temporal datasets.The proposed MSFCN achieves 60.366%on the WHDLD dataset and 75.127%on the GID dataset in terms of mIoU index while the figures for two spatio-temporal datasets are 87.753%and 77.156%.Extensive comparative experiments and abla-tion studies demonstrate the effectiveness of the proposed MSFCN.
文摘This paper aims at providing multi-source remote sensing images registered in geometric space for image fusion.Focusing on the characteristics and differences of multi-source remote sensing images,a feature-based registration algorithm is implemented.The key technologies include image scale-space for implementing multi-scale properties,Harris corner detection for keypoints extraction,and partial intensity invariant feature descriptor(PIIFD)for keypoints description.Eventually,a multi-scale Harris-PIIFD image registration algorithm framework is proposed.The experimental results of fifteen sets of representative real data show that the algorithm has excellent,stable performance in multi-source remote sensing image registration,and can achieve accurate spatial alignment,which has strong practical application value and certain generalization ability.
基金This project is supported by the National Natural Science Foundation of China(NSFC)(No.61902158).
文摘The degradation of optical remote sensing images due to atmospheric haze poses a significant obstacle,profoundly impeding their effective utilization across various domains.Dehazing methodologies have emerged as pivotal components of image preprocessing,fostering an improvement in the quality of remote sensing imagery.This enhancement renders remote sensing data more indispensable,thereby enhancing the accuracy of target iden-tification.Conventional defogging techniques based on simplistic atmospheric degradation models have proven inadequate for mitigating non-uniform haze within remotely sensed images.In response to this challenge,a novel UNet Residual Attention Network(URA-Net)is proposed.This paradigmatic approach materializes as an end-to-end convolutional neural network distinguished by its utilization of multi-scale dense feature fusion clusters and gated jump connections.The essence of our methodology lies in local feature fusion within dense residual clusters,enabling the extraction of pertinent features from both preceding and current local data,depending on contextual demands.The intelligently orchestrated gated structures facilitate the propagation of these features to the decoder,resulting in superior outcomes in haze removal.Empirical validation through a plethora of experiments substantiates the efficacy of URA-Net,demonstrating its superior performance compared to existing methods when applied to established datasets for remote sensing image defogging.On the RICE-1 dataset,URA-Net achieves a Peak Signal-to-Noise Ratio(PSNR)of 29.07 dB,surpassing the Dark Channel Prior(DCP)by 11.17 dB,the All-in-One Network for Dehazing(AOD)by 7.82 dB,the Optimal Transmission Map and Adaptive Atmospheric Light For Dehazing(OTM-AAL)by 5.37 dB,the Unsupervised Single Image Dehazing(USID)by 8.0 dB,and the Superpixel-based Remote Sensing Image Dehazing(SRD)by 8.5 dB.Particularly noteworthy,on the SateHaze1k dataset,URA-Net attains preeminence in overall performance,yielding defogged images characterized by consistent visual quality.This underscores the contribution of the research to the advancement of remote sensing technology,providing a robust and efficient solution for alleviating the adverse effects of haze on image quality.
基金supported by the National Natural Science Foundation of China(Nos.42371094,41907253)partially supported by the Interdisciplinary Cultivation Program of Xidian University(No.21103240005)the Postdoctoral Fellowship Program of CPSF(No.GZB20240589)。
文摘INTRODUCTION.On May 1st,2024,around 2:10 a.m.,a catastrophic collapse occurred along the Meilong Expressway near Meizhou City,Guangdong Province,China,at coordinates 24°29′24″N and 116°40′25″E.This collapse resulted in a pavement failure of approximately 17.9 m in length and covering an area of about 184.3 m^(2)(Chinanews,2024).
文摘With the increasing global population and mounting pressures on agricultural production,precise pest monitoring has become a critical factor in ensuring food security.Traditional monitoring methods,often inefficient,struggle to meet the demands of modern agriculture.Drone remote sensing technology,leveraging its high efficiency and flexibility,demonstrates significant potential in pest monitoring.Equipped with multispectral,hyperspectral,and thermal infrared sensors,drones can rapidly cover large agricultural fields,capturing high-resolution imagery and data to detect spectral variations in crops.This enables effective differentiation between healthy and infested plants,facilitating early pest identification and targeted control.This paper systematically reviews the current applications of drone remote sensing technology in pest monitoring by examining different sensor types and their use in monitoring major crop pests and diseases.It also discusses existing challenges,aiming to provide insights and references for future research.
基金supported by the National Natural Science Foundation of China(42030102,42371321).
文摘Accurate fine-grained geospatial scene classification using remote sensing imagery is essential for a wide range of applications.However,existing approaches often rely on manually zooming remote sensing images at different scales to create typical scene samples.This approach fails to adequately support the fixed-resolution image interpretation requirements in real-world scenarios.To address this limitation,we introduce the million-scale fine-grained geospatial scene classification dataset(MEET),which contains over 1.03 million zoom-free remote sensing scene samples,manually annotated into 80 fine-grained categories.In MEET,each scene sample follows a scene-in-scene layout,where the central scene serves as the reference,and auxiliary scenes provide crucial spatial context for fine-grained classification.Moreover,to tackle the emerging challenge of scene-in-scene classification,we present the context-aware transformer(CAT),a model specifically designed for this task,which adaptively fuses spatial context to accurately classify the scene samples.CAT adaptively fuses spatial context to accurately classify the scene samples by learning attentional features that capture the relationships between the center and auxiliary scenes.Based on MEET,we establish a comprehensive benchmark for fine-grained geospatial scene classification,evaluating CAT against 11 competitive baselines.The results demonstrate that CAT significantly outperforms these baselines,achieving a 1.88%higher balanced accuracy(BA)with the Swin-Large backbone,and a notable 7.87%improvement with the Swin-Huge backbone.Further experiments validate the effectiveness of each module in CAT and show the practical applicability of CAT in the urban functional zone mapping.The source code and dataset will be publicly available at https://jerrywyn.github.io/project/MEET.html.
基金supported by National Key R&D Program of China(2022YFD2000100)National Natural Science Foundation of China(42401400)Zhejiang Provincial Key Research and Development Program(2023C02018).
文摘Yellow rust(Puccinia striiformis f.sp.Tritici,YR)and fusarium head blight(Fusarium graminearum,FHB)are the two main diseases affecting wheat in the main grain-producing areas of East China,which is common for the two diseases to appear simultaneously in some main production areas.It is necessary to discriminate wheat YR and FHB at the regional scale to accurately locate the disease in space,conduct detailed disease severity monitoring,and scientific control.Four images on different dates were acquired from Sentinel-2,Landsat-8,and Gaofen-1 during the critical period of winter wheat,and 22 remote sensing features that characterize the wheat growth status were then calculated.Meanwhile,6 meteorological parameters that reflect the wheat phenological information were also obtained by combining the site meteorological data and spatial interpolation technology.Then,the principal components(PCs)of comprehensive remote sensing and meteorological features were extracted with principal component analysis(PCA).The PCs-based discrimination models were established to map YR and FHB damage using the random forest(RF)and backpropagation neural network(BPNN).The models’performance was verified based on the disease field truth data(57 plots during the filling period)and 5-fold cross-validation.The results revealed that the PCs obtained after PCA dimensionality reduction outperformed the initial features(IFs)from remote sensing and meteorology in discriminating between the two diseases.Compared to the IFs,the average area under the curve for both micro-average and macro-average ROC curves increased by 0.07 in the PCs-based RF models and increased by 0.16 and 0.13,respectively,in the PCs-based BPNN models.Notably,the PCs-based BPNN discrimination model emerged as the most effective,achieving an overall accuracy of 83.9%.Our proposed discrimination model for wheat YR and FHB,coupled with multi-source remote sensing images and meteorological data,overcomes the limitations of a single-sensor and single-phase remote sensing information in multiple stress discrimination in cloudy and rainy areas.It performs well in revealing the damage spatial distribution of the two diseases at a regional scale,providing a basis for detailed disease severity monitoring,and scientific prevention and control.
基金Supported by the National Natural Science Foundation of China(42401488,42071351)the National Key Research and Development Program of China(2020YFA0608501,2017YFB0504204)+4 种基金the Liaoning Revitalization Talents Program(XLYC1802027)the Talent Recruited Program of the Chinese Academy of Science(Y938091)the Project Supported Discipline Innovation Team of the Liaoning Technical University(LNTU20TD-23)the Liaoning Province Doctoral Research Initiation Fund Program(2023-BS-202)the Basic Research Projects of Liaoning Department of Education(JYTQN2023202)。
文摘Accurate estimation of understory terrain has significant scientific importance for maintaining ecosystem balance and biodiversity conservation.Addressing the issue of inadequate representation of spatial heterogeneity when traditional forest topographic inversion methods consider the entire forest as the inversion unit,this study pro⁃poses a differentiated modeling approach to forest types based on refined land cover classification.Taking Puerto Ri⁃co and Maryland as study areas,a multi-dimensional feature system is constructed by integrating multi-source re⁃mote sensing data:ICESat-2 spaceborne LiDAR is used to obtain benchmark values for understory terrain,topo⁃graphic factors such as slope and aspect are extracted based on SRTM data,and vegetation cover characteristics are analyzed using Landsat-8 multispectral imagery.This study incorporates forest type as a classification modeling con⁃dition and applies the random forest algorithm to build differentiated topographic inversion models.Experimental re⁃sults indicate that,compared to traditional whole-area modeling methods(RMSE=5.06 m),forest type-based classi⁃fication modeling significantly improves the accuracy of understory terrain estimation(RMSE=2.94 m),validating the effectiveness of spatial heterogeneity modeling.Further sensitivity analysis reveals that canopy structure parame⁃ters(with RMSE variation reaching 4.11 m)exert a stronger regulatory effect on estimation accuracy compared to forest cover,providing important theoretical support for optimizing remote sensing models of forest topography.
文摘Cloud detection is a critical preprocessing step in remote sensing image processing, as the presence of clouds significantly affects the accuracy of remote sensing data and limits its applicability across various domains. This study presents an enhanced cloud detection method based on the U-Net architecture, designed to address the challenges of multi-scale cloud features and long-range dependencies inherent in remote sensing imagery. A Multi-Scale Dilated Attention (MSDA) module is introduced to effectively integrate multi-scale information and model long-range dependencies across different scales, enhancing the model’s ability to detect clouds of varying sizes. Additionally, a Multi-Head Self-Attention (MHSA) mechanism is incorporated to improve the model’s capacity for capturing finer details, particularly in distinguishing thin clouds from surface features. A multi-path supervision mechanism is also devised to ensure the model learns cloud features at multiple scales, further boosting the accuracy and robustness of cloud mask generation. Experimental results demonstrate that the enhanced model achieves superior performance compared to other benchmarked methods in complex scenarios. It significantly improves cloud detection accuracy, highlighting its strong potential for practical applications in cloud detection tasks.
基金supported by Henan Province Key R&D Project(241111210400)Henan Provincial Science and Technology Research Project(242102211007 and 242102211020)+1 种基金Jiangsu Science and Technology Programme-General Programme(BK20221260)Science and Technology Innovation Project of Zhengzhou University of Light Industry(23XNKJTD0205).
文摘In recent years,convolutional neural networks(CNN)and Transformer architectures have made significant progress in the field of remote sensing(RS)change detection(CD).Most of the existing methods directly stack multiple layers of Transformer blocks,which achieves considerable improvement in capturing variations,but at a rather high computational cost.We propose a channel-Efficient Change Detection Network(CE-CDNet)to address the problems of high computational cost and imbalanced detection accuracy in remote sensing building change detection.The adaptive multi-scale feature fusion module(CAMSF)and lightweight Transformer decoder(LTD)are introduced to improve the change detection effect.The CAMSF module can adaptively fuse multi-scale features to improve the model’s ability to detect building changes in complex scenes.In addition,the LTD module reduces computational costs and maintains high detection accuracy through an optimized self-attention mechanism and dimensionality reduction operation.Experimental test results on three commonly used remote sensing building change detection data sets show that CE-CDNet can reduce a certain amount of computational overhead while maintaining detection accuracy comparable to existing mainstream models,showing good performance advantages.
